The Fabric Of Geology

1 025

The Fabric o

Reading, Massachusetts Palo Alto London

CLAUDE C. ALBRITTON, Jr., Editor

f GEOLOGY

Prepared under the direction of a Committee
of The Geological Society of America, in
commemoration of the Society's 75th Anniversary

ADDISON-WESLEY PUBLISHING COMPANY, INC.

Copyright 1963
GEOLOGICAL SOCIETY OF AMERICA

Printed in the United States of America

ALL RIGHTS RESERVED. THIS BOOK, OR PARTS

THEREOF, MAY NOT BE REPRODUCED IN ANY FORM

WITHOUT WRITTEN PERMISSION OF THE SOCIETY.

Library of Congress Catalog Card No. 63-77726

Preface

During the years immediately following World War II, many geologists
became dissatisfied with the training that was being given to students of the
earth sciences. Accordingly, the Council of The Geological Society of America,
in December of 1946, named a committee to investigate the state of geologic
education, and to offer suggestions for its improvement. The report of this
committee appeared in the Interim Proceedings of the Society for 1949. Of the
recommendations presented, the one emphasized most strongly urged that, at
all levels of instruction, "only those inferences be presented ... for which the
essential observational data and the logical steps leading to the inference have
also been presented."* Before this criterion could be satisfied, the committee-
men asserted, the logical structure of geologic science would have to be re-
examined "from the ground up."

When the time came to make plans for the Society's seventy-fifth anniversary,
the Councilors, recalling this last recommendation, decided that the theme for
the anniversary meetings should be the philosophy of geology. A committee
was asked to produce a book of essays on the fabric of geologic thought, and to
arrange a program on the same subject for the annual meetings in 1963.

The very lack of any modern book on the philosophy of geology is justifica-
tion enough for this work. The members of the Anniversary Committee will
have achieved their purpose if this collection, despite any shortcomings it may
have, serves as a focal point for discussions of our role as scientists.

The book begins with a toast to James Hutton, as founder of modern geology.
Mclntyre seeks out the origin of the ideas which shaped Hutton's fruitful theory
of the earth, and he finds some likely sources in some unlikely places including
steam engines, organisms, and the all but forgotten work of George Hoggart
Toulmin.

Does geology have laws and theories of its own? What is to be understood
when geology is called historical science? These questions are considered in a
sequence of three essays. Bradley identifies the history and constitution of the
earth as the two principal subjects of geologic investigation. Simpson goes on
to develop the differences between the historical and the nonhistorical aspects
of the science, which he finds in their respective concerns with the configurational
as opposed to the immanent properties of matter and energy. If "laws," in the

* Hubbert, M. K., Hendricks, T. A., and Thiel, G. A. (Chairman), Report of the
Committee on Geologic Education of The Geological Society of America: Geol. Soc.
Am., Int. Pr., 1949, pt. 2, pp. 17-21.

vi PREFACE

usual sense of this term, apply to the immanent and not to the configurational
in nature, then the laws of physical and dynamic geology are the laws of physics
and chemistry. Kitts concludes that the theory of geology is likewise the theory
of physics and chemistry used by geologists, however, as instruments of his-
torical inference.

Laboratory experiments have played a less important role in geology than in
chemistry or physics. And yet, as H. H. Read has observed, "Any man looking
out of any window sees a geological laboratory in constant and full-scale opera-
tion." McKelvey develops this theme by showing how geologists have the un-
usual opportunity to observe the results of complex natural experiments that
involve larger masses and longer periods of time than can be handled in the
laboratory.

The six essays forming the middle section of the book treat of geologic thought
within the framework of some particular branch of earth science. Woodford
critically examines the proposition in stratigraphy that fossils may be used to
order strata in chronologic sequences. Mclntyre investigates the reliability of
certain methods for obtaining the absolute ages of rocks that are used to cali-
brate the stratigraphic column described by Woodford. Mclntyre's essay also
illustrates a trend toward the quantitative in geology, and this is the theme of
Mackin's comparison of the rational and empirical methods of investigation.
Mackin draws his examples from geomorphology, the same field which Leopold
and Langbein use to illustrate the association of ideas in geologic thought.
Structural geology provides Hill with an example which he uses to develop
rules of geologic nomenclature and classification. Anderson also turns to
structural geology for examples to show how the geologist uses the logical
principle of simplicity.

Three essays are concerned primarily with the communication of geologic
data and ideas. Betz examines the documentary tools of communication and
suggests how these might be better used to contend with the "information
problem" which is of growing concern to all scientists. Gilluly shows how
G. K. Gilbert, in his scientific memoirs, took pains to disclose the chains of
reasoning that linked his observations to his conclusions. Harrison discusses
the interpretative character of the geologic map, and illustrates the way in
which theory may influence mapping, and vice versa.

The last two essays in the book identify contrasting areas of geologic thought
and work which are in need of further development. Hagner urges that more
attention be given to the history and philosophy of geology, while Legget points
to the increasing opportunities of putting geology to work in the service of man.

A bibliography of writings that reflect upon the character of geologic thought
concludes the volume.

The planning of this book has been the resonsibility of a committee of eight
persons, who, in addition to the Editor, are Messrs. Frederick Betz, Jr., James

PREFACE vii

Gilluly, J. M. Harrison, H. H. Hess, Mason L. Hill, Luna B. Leopold, and
W. W. Rubey.

The members of the Anniversary Committee thank the Trustees of The
Graduate Research Center, Inc. for sponsoring a conference on the scope and
philosophy of geology, which was held in Dallas in October of 1960. At this
meeting, membeis of the Committee presented their tentative thoughts on the
logical foundations of geology. Three philosophers Nelson Goodman, Carl G.
Hempel and John H. Kultgen served as critics, and were especially helpful in
drawing lines between real problems and false issues. William E. Benson and
Eugene Herrin were guests of the conference and active participants in discus-
sions which did much to shape the contents of the book.

The idea of bringing philosophers and geologists together for the conference
in Dallas belongs to Mr. W. H. Freeman. As the Publisher's Editor for this
book, he has, moreover, arranged to have most of the essays read by Professors
Morton Beckner and Michael Scriven, whose forthright criticisms are grate-
fully acknowledged.

Mrs. Robert R. Wheeler prepared the special index for the annotated bibli-
ography and also the general index for this book. Mrs. Jacquelyn Newbury
typed much of the manuscript and obtained permission to reproduce the
numerous quotations in the bibliography and text.

"When 'Omer Smote 'is Bloomin' Lyre," by Rudyard Kipling, is reprinted
from Rudyard Kipling's Verse (Definitive Edition) by permission of Mrs. George
Bambridge and Doubleday & Company, Inc.

Dallas, Texas C. C. A.

July 7, 1963

Contents

James Hutton and the philosophy of geology 1

DONALD B. MC!NTYRE

Geologic laws 12

W. H. BRADLEY

Historical science 24

GEORGE GAYLORD SIMPSON

The theory of geology 49

DAVID B. Krrrs

Geology as the study of complex natural experiments 69

V. E. McKELVEY

Correlation by fossils 75

A. O. WOODFORD

Precision and resolution hi geochronometry 112

DONALD B. MC!NTYRE

Rational and empirical methods of investigation in geology 135

J. HOOVER MACKIN

Role of classification in geology 164

MASON L. HILT.

Simplicity in structural geology 175

CHARLES A. ANDERSON

Association and indeterminacy in geomorphology 184

LUNA B. LEOPOLD AND WALTER B. LANGBEIN
ix

x CONTENTS

Geologic communication 193

FREDERICK BETZ, JR.

The scientific philosophy of G. K. Gilbert 218

JAMES GILLULY

Nature and significance of geological maps 225

M. M. HARRISON

Philosophical aspects of the geological sciences 233

ARTHUR F. HAGNER

Geology in the service of man 242

ROBERT F. LEGGET

Philosophy of geology: A selected bibliography and index 262

CLAUDE C. ALBRTTTON, JR.

Index 367

DONALD B. McINTYRE

Pomona College

James Hutton and the

Philosophy of Geology 1

"When 'Omer smote 'is bloomin' lyre,
He'd 'card men sing by land an' sea;

An' what he thought 'e might require,
'E went an' took the same as me!

"The market-girls an' fishermen,

The shepherds an' the sailors, too,

They 'card old songs turn up again,
But kep' it quiet same as you!

"They knew 'e stole; 'e knew they knowed.

They didn't tell, nor make a fuss,
But winked at 'Omer down the road,

An' 'e winked back the same as us!"

KIPLING

1 On the 150th anniversary of Button's death it was my good fortune to attend, in
Edinburgh, a memorable lecture by S. I. Tomkeieff. The effect it had on me was pro-
found and, because my present subject concerns the unavoidable and indeed laudable
practice of borrowing of ideas, I have not hesitated to adopt TomkeiefPs title and a
little more besides. The extent of my indebtedness may be gauged by reference to
TomkeiefPs paper (1948).

1

2 DONALD B. McINTYRE

Prior to Hutton, geology did not exist, and I think it is generally agreed
that the science was created in the fifty years between 1775 and 1825. In
1775, Hutton's close friend, James Watt, constructed his first working steam
engine, and Werner, who was to become the leader of the Neptunists, began
teaching at the Mining Academy of Freiburg. By 1825, the first passenger
train was in operation, and Lyell, who was born the year Hutton died, was
at work on his "Principles of Geology" "the coping stone" of the new science.
The same years saw the American War of Independence, the Industrial Revo-
lution, the French Revolution, and the Napoleonic wars. It was the age of
Wordsworth and Coleridge, Byron and Shelley, Goethe and Schiller, Scott
and Burns.

During this period the necessary foundations for petrology, and the study of
the materials of the crust, were laid by the work of Haiiy on crystallography,
Werner on the physical properties of minerals, and Joseph Black on their
chemical composition. Stratigraphical paleontology and the technique of field
mapping were initiated by William Smith, but it was Hutton, and Hutton
alone, who provided geology with a dynamic scheme a theory, in the original
sense of "something seen in the mind." Thus Hutton plays the same role in
geology as Newton in astronomy, or Darwin in biology. Comprehension is
the power of the mind to understand, and these intellectual giants have given
us comprehension of the great processes which go on around us. For this
reason, Hutton truly deserves the title "Founder of Modern Geology."

Fortunately for posterity, Hutton had a most eloquent biographer in John
Playfair, Professor of Physics and of Mathematics at the University of Edin-
burgh, and one of the finest writers of scientific prose the English language has
known. If the history of science is to be the study of the origin and development
of ideas, then the history of geology needs to give attention to Playfair's re-
mark: "It would be desirable to trace the progress of Dr. Hutton's mind in the
formation of a system where so many new and enlarged views of nature occur,
and where so much originality is displayed."

But how can anyone, other than the author himself, possibly hope to follow
the sequence of ideas that unconsciously have been at work in a creative
mind? The classic answer to this question has been given by John Livingston
Lowes in his study of the process of creation in the mind of Coleridge, entitled
"The Road to Xanadu" one of the most remarkable works of detection ever
written. Coleridge kept a notebook in which he recorded facts and phrases
which had caught his attention, and Lowes was able to use this book, along
with library records, to follow Coleridge through his reading and hence "to
retrace," as he said, "the obliterated vestiges of creation." The result is aston-
ishing: the origin of almost every phrase and analogy in "The Rime of the
Ancient Mariner" can be identified; and quite an assortment these sources are,

JAMES HUTTON AND THE PHILOSOPHY OF GEOLOGY 3

including as they do such unlikely works as Burnett's "Sacred Theory of the
Earth," Maupertuis' geodetic study on "The Figure of the Earth, 9 ' Father
Bourges* "Luminous Appearances in the Wakes of Ships," and the Astronomer
Royal's paper in the Philosophical Transactions of the Royal Society entitled
"An account of an appearance, like a star, seen in the dark part of the moon."

Hutton, we are told, "was in the habit of using his pen continually as an
instrument of thought," and he left behind him what is described as "an
incredible quantity of manuscript"; unhappily almost none has survived, so
we are deprived of the clues his notebooks would have afforded.

Coleridge himself was keenly aware ot the pattern: "the imagination, the
true inward creatrix," constantly working on "the shattered fragments of
memory, dissolves, diffuses, dissipates, in order to recreate." And elsewhere
he wrote that the "hooks-and-eyes of memory permit a confluence of our recol-
lections." The analogy is reminiscent of Poincar6's image of mathematical
discovery, in which he pictures ideas as hooked atoms ploughing through space
like the molecules as conceived in the kinetic theory of gases. The number of
possible combinations is so vast that even in a lifetime they couldn't all be
examined, and PoincarS concluded that esthetic sensibility, the appreciation
of the beauty of an elegant solution, acts in the subconscious like a delicate
sieve to catch combinations worthy of the attention of the conscious mind.
Thus, the creation of a poem and the discovery of a mathematical law may
have much hi common. Perhaps a geologic theory also is born in the same
manner.

A long time ago that giant of American geology, Grove Karl Gilbert,
elaborated this theme in two masterly presidential addresses: one was entitled
"The Inculcation of Scientific Method by Example" (1885) and the other
"The Origin of Hypotheses" (1895). Gilbert, himself no mean deviser of
brilliant and ingenious ideas, made a practice of trying to analyze the methods
he and his associates used in geologic research.

He wrote: "Just as in the domain of matter nothing is created from nothing,
just as in the domain of life there is no spontaneous generation, so in the
domain of mind there are no ideas which do not owe their existence to ante-
cedent ideas which stand in the relation of parent to child. It is only because
our mental processes are largely conducted outside of consciousness that the
lineage of ideas is difficult to trace ... To explain the origin of hypotheses"
wrote Gilbert, "I have a hypothesis to present. It is that hypotheses are
always suggested through analogy. Consequential relations of nature are in-
finite in variety, and he who is acquainted with the largest number has the
broadest base for the analogic suggestion of hypotheses." I believe it was
H. H. Read who affirmed that "the best geologist is he who has seen the most
geology.* 1 At any rate, there is undoubtedly some truth in the remark.

4 DONALD B. McINTYRE

The conclusion we reach seems to be that the equipment necessary to devise
hypotheses includes:

First, an excellent memory and an extensive knowledge of the relevant
material;

Second, the ability to form associations of ideas, and to reason by analogy;
and

Third, the possession of an unusually high degree of esthetic feeling for an
elegant solution, and a burning enthusiasm for the subject.

Now Playfair (1805, p. 98) informs us that Dr. Hutton:

. . . had acquired great information; and an excellent memory supplied an
inexhaustible fund of illustration, always happily introduced. He used also
regularly to unbend himself with a few friends, in the little society . . . known
by the name of the Oyster Club. . . The original members of it were Mr.
[Adam] Smith [the author of "The Wealth of Nations"], Dr. [Joseph] Black
[the discoverer of latent heat and of carbon dioxide, and who introduced
quantitative methods in chemistry], and Dr. Hutton, and round them was
soon formed a knot of those who knew how to value the familiar and social
converse of these illustrious men. As all the three possessed great talents,
enlarged views, and extensive information, without any of the stateliness
and formality which men of letters think it sometimes necessary to affect;
as they were all three easily amused, were equally prepared to speak and to
listen, and as the sincerity of their friendship had never been darkened by
the least shade of envy; it would be hard to find an example where every
thing favourable to good society was more perfectly united, and every thing
adverse more entirely excluded. The conversation was always free, often
scientific, but never didactic or disputatious; and as this club was much the
resort of the strangers who visited Edinburgh, from any object connected
with art or with science, it derived from thence an extraordinary degree of
variety and interest.

Hutton, then, was peculiarly well placed for the accumulation of diverse
information and ideas, and he had an excellent memory. What of his ability
to reason by analogy? Playfair records that Hutton possessed "the experienced
eye, the power of perceiving the minute differences and fine analogies which
discriminate or unite the objects of science, and the readiness of comparing
new phenomena with others already treasured up in the mind."

What then of his appreciation of the beauty of an elegant solution, and what
of his enthusiasm?

A circumstance which greatly distinguished the intellectual character of
the philosopher of whom we now speak, was an uncommon activity and ar-
dour of mind, upheld by the greatest admiration of whatever in science was

JAMES HUTTON AND THE PHILOSOPHY OF GEOLOGY 5

new, beautiful, or sublime. The acquisitions of fortune, and the enjoyments
which most directly address the senses, do not call up more lively expressions
of joy in other men, than hearing of a new invention, or being made ac-
quainted with a new truth, would, at any time, do in Dr. Hutton. This
sensibility to intellectual pleasure was not confined to a few objects, nor to
the sciences which he particularly cultivated: he would rejoice over Watt's
improvements on the steam engine, or Cook's discoveries in the South Sea,
with all the warmth of a man who was to share in the honour or the profit
about to accrue from them. The fire of his expression on such occasions,
and the animation of his countenance and manner, are not to be described;
they were always seen with great delight by those who could enter into his
sentiments, and often with great astonishment by those who could not.

With this exquisite relish for whatever is beautiful and sublime in science,
we may easily conceive what pleasure he derived from his own geological
speculations. The novelty and grandeur of the objects offered by them to
the imagination, the simple and uniform order given to the whole natural
history of the earth, . . . are things to which hardly any man could be in-
sensible; but to him they were matter, not of transient delight, but of solid
and permanent happiness. (Playfair, 1805, p. 91)

Elsewhere Playfair remarks that both Hutton and Black were "formed with
a taste for what is beautiful and great in science, with minds inventive and
fertile in new combinations." Clearly, Hutton had all the qualities we have
suggested as necessary in a great scientific synthesizer and he possessed them
to a remarkable degree. We must now look at the most likely sources of his
ideas: his immediate background and the achievements of his friends.

Hutton' s first serious studies were in chemistry and medicine, first in Edin-
burgh, then in Paris, and finally in Leyden, where he took his M.D. degree
at the age of 23. However, he never practiced medicine, for, on his return to
Scotland, he took charge of the small farm which he had inherited from his
father. Hutton, who never did things by halves, immediately applied his
scientific training to agriculture. He introduced new methods to Scottish
farming, and he traveled to Norfolk and the Low Countries in search of the
best techniques and practices. While on these journeys he became increasingly
interested in the origin of soil and in the processes of geology.

After 13 years of successful farming, he moved to Edinburgh, where his prin-
cipal income was from his ammonium chloride plant the first in Britain.
His time was spent in reading, in chemical experiments, often together with
Black, and in the company of his illustrious and stimulating friends. He was
a member of the Council of the newly organized Royal Society of Edinburgh,
and for the first volume of the Society's Transactions he prepared his paper
on the "Theory of the Earth, or an investigation of the laws observable in the
composition, dissolution, and restoration of land upon the globe." The paper
was read at two successive meetings of the Society in 1785. Illness prevented

6 DONALD B. McINTYRE

Hutton from attending the first of these meetings, and the first part of his paper
was read by his friend Joseph Black.

In addition to his studies of medicine, agriculture, chemistry, meteorology,
and geology, he published a three-volume work on philosophy and a disserta-
tion on the Chinese language.

The accomplishments of his friends are so extensive that we can afford only
a catalog. His intimate acquaintances were responsible for the discovery of
carbon dioxide, nitrogen, oxygen, and strontium, and he himself was the first
to extract sodium from a silicate. The list includes the discovery that water
has a point of maximum density and that latent heat is needed to change its
state. His friends included men responsible for the development of the steam
engine; for the founding of iron works and a sulfuric acid plant; for the use of
chlorine in bleaching; the author of the first book on agricultural chemistry;
the author of the "Wealth of Nations"; the man to whom Sir Walter Scott
dedicated "Waverley"; and that remarkable man who, never having been to
sea, devised a system of naval tactics which won the British fleet several vic-
tories, and which was quoted in Nelson's battle orders at Trafalgar.

With this background we are now equipped to examine Hutton's Theory,
and we will also draw somewhat on the unpublished manuscript of his "Ele-
ments of Agriculture" on which he was working at the time of his death.
Insofar as is practicable, I will give you Hutton's own words.

First of all we need to know that for Hutton, "A theory is nothing but the
generalization of particular facts; and, in a theory of the earth, those facts must
be taken from the observations of natural history." For Hutton, a phenomenon
(what he termed an "appearance") was "explained" when it had been "com-
prehended" by a theory, that is to say, incorporated into the structure of the
theory.

Second, Hutton remarked that "It is with pleasure that man observes order
and regularity in the works of nature, instead of becoming disgusted with
disorder and confusion. If the stone which fell today were to rise tomorrow,
there would be an end of natural philosophy, our principles would fail, and
we would no longer investigate the rules of nature from our observations."

This, of course, is the doctrine of uniformitarianism, but it wasn't original
with Hutton. Before Hutton went to Leyden, the Professor of Astronomy there
wrote: "When, as a result of certain observations, we anticipate other cases
which we have not directly observed, our prediction is based on the axiom of
uniformity of nature. All action- would be impossibly if we could not assume
that the lessons of former experience would be valid in the future."

For Hutton it was clear that, in his own words, "We must read the trans-
actions of time past in the present state of natural bodies, and, for the reading
of this character, we have nothing but the laws of nature, established in the
science of man by his inductive reasoning. For man is not satisfied in seeing

JAMES HUTTON AND THE PHILOSOPHY OF GEOLOGY 7

things which are; he seeks to know how things have been, and what they are
to be." However, it was Sir Archibald Geikie and not James Hutton who
crystallized this concept in the memorable dictum: "The present is the key to
the past. 35

In Hutton's own opinion, it was agriculture that had been the study of his
life; geology had been incidental. Accordingly he looked on the world as a
well-run farm, designed to sustain plants and animals, and with rotation,
necessary to maintain fertility. Indeed, the secret of Hutton is that he thought
of the world as a sort of superorganism. His was not the mind of a narrow
specialist. For him the biological sciences were completely integrated with the
physical. "Here," he said, "is a compound system of things, forming together
one whole living world."

The most solid rocks moulder and decay upon the surface of the earth,
and thus procure a soil, either immediately upon the place which, thus, had
given it birth, or remotely upon some other place where it may be trans-
ported by the water or the wind. For this great purpose of the world, the
solid structure of the earth must be sacrificed; for the fertility of the soil
depends upon the loose and incoherent state of its materials; and this state
of the fertile soil necessarily exposes it to the ravages of the rain upon the
inclined surface of the earth.

From the tops of the mountains to the shores of the sea, all the soils are
subject to be moved from their places, and to be deposited in a lower situa-
tion; thus gradually proceeding from the mountain to the river; and from
the river, step by step, into the sea. If the vegetable soil is thus constantly
removed from the surface of the land, and if its place is thus to be supplied
from the dissolution of the solid earth, ... we may perceive an end to this
beautiful machine; an end arising from . . . that destructibility of its land
which is so necessary in the system of the globe, in the economy of life and
vegetation. It may be concluded that the apparent permanency of 'this
earth is not real or absolute, and that the fertility of its surface, like the
healthy state of animal bodies, must have its period and be succeeded by
another. (Hutton, 1788, p. 215 et passim)

"We have now considered the globe of this earth as a machine, constructed
upon chemical as well as mechanical principles, . . . But is this world [he asks]
to be considered thus merely as a machine, to last no longer than its parts
retain their present position, their proper forms and qualities?" And here we
see Watt and his steam engine in Hutton's mind. "Or, [he asks] may it not
be also considered as an organized body such as has a constitution in which
the necessary decay of the machine is naturally repaired . . . [Is] there, . . .
in the constitution of this world, a reproductive operation by which a ruined
constitution may be again repaired?" (Hutton, 1788, p. 215) And here we
see the physician and the farmer.

8 DONALD B. McINTYRE

"From the constitution of those materials which compose the present land,
we have reason to conclude that, during the time this land was forming, by the
collection of its materials at the bottom of the sea, there had been a former
land containing minerals similar to those we find at present in examining the
earth ... A habitable earth is made to rise out of the wreck of a former world."
And this is Hutton the geologist.

The whole spirit of Mutton's geology is contained in his statement that
"The matter of this active world is perpetually moved, in that salutary circu-
lation [a good medical expression!] by which provision is so wisely made for
the growth and prosperity of plants, and for the life and comfort of the various
animals."

Now it is a most remarkable fact that, immediately prior to the publication
of Hutton' s Theory in 1788, there was a man, today almost completely for-
gotten, who viewed the earth just as Hutton did. His name was George
Hoggart Toulmin. No one familiar with Hutton's writings can read Toulmin's
"The Antiquity of the World" without being impressed by astonishing simi-
larities. For my own part, I find it impossible to avoid the conclusion that
Hutton had read it prior to writing his own paper.

Like Hutton, Toulmin took a remarkably comprehensive view of the earth,
referring to "the beautiful order and disposition of the several parts that com-
pose the stupendous whole." Like Hutton, he adopted a fundamental uni-
formitarianism: "Nature is always the same, her laws are eternal and im-
mutable." Like Hutton he believed that slow changes, long continued, can
produce far-reaching results: "These immutable truths should never be for-
got," he writes, "that animals and vegetables flourish and decay; that earths
are formed by slow degrees; that they too change by time; that stone is formed,
is decomposed or altered in its composition; that mountains now are elevated;
now depressed; that nature lives in motion."

Like Hutton, he recognized the significance of sedimentary rocks as proof
of the circulation of matter. But to me it was of particular interest that Toul-
min's very words and phrases echoed what I knew in Hutton: "The continual
formation and decay of every existing substance, the unceasing circulation of
matter, produces no disorder. A continual waste in every part is necessary to
the incessant repairs of the whole. The closest sympathy and connection is
preserved throughout the entire system of things."

Now Hutton wrote: We are "led to acknowledge an order in a subject which,
in another view, has appeared as absolute disorder and confusion. . . There
is a certain order established for the progress of nature, for the succession of
things, and for the circulation of matter upon the surface of the globe. . .
We must see how this machine is so contrived as to have those parts which are
wasting and decaying, again repaired ... the necessary decay is naturally
repaired."

JAMES HUTTON AND THE PHILOSOPHY OF GEOLOGY 9

"We are," says Hutton, "thus led to see a circulation in the matter of the
globe, and a system of beautiful economy in the works of nature. This earth,
like the body of an animal, is wasted at the same time that it is repaired. It
has a state of growth and augmentation; it has another state which is that of
diminution and decay. This world is thus destroyed in one part, but it is
renewed in another."

The last sentence of Toulmin's book ends: "We have by no means been led
to contravene . . . the existence of infinite intelligence and wisdom." And else-
where he refers to "nature, whose every operation is stamped with wisdom and
consistency." The last paragraph of Hutton's paper begins, "We have now
got to the end of our reasoning; we have the satisfaction to find that in nature
there is wisdom, system, and consistency."

Toulmin wrote: "there has ever been a succession of events, something
similar to what is continually observed ... a vast succession of ages. . . We
have been induced to conclude that the whole system of things," and so on.
Hutton put it thus: "having seen a succession of worlds, we may from this
conclude that there is a system in nature."

Toulmin said: "In the circle of existence, in vain do we seek the beginning
of things." And Hutton, intimating that he had reached the limit of his
vision into the past, wrote: "It is in vain to look for anything higher in the
origin of the earth."

Toulmin's main point is that "through the whole of this enquiry we have
endeavoured to demonstrate . . . that, as there never was any beginning, so
will there never be a conclusion. . ." And he repeats, "the whole system of
things never had any beginning, nor will have any termination." And the
final sentence of Hutton's paper is: "The result, therefore, of our present
enquiry is, that we find no vestige of a beginning, no prospect of an end."
Are we not, like John Livingston Lowes, "retracing the obliterated vestiges of
creation"? 2

In 1749 Hutton was granted the M.D. degree at Leyden for his thesis on
"The Blood and Circulation in the Microcosm." The title reminds us that,
from the beginning of speculative thought, philosophers have pondered the
concept that man, the microcosm, is the epitome of the macrocosm or the
world in which he lives. Attempts were constantly being made to find analogies
or correspondences between aspects of the anatomy and physiology of man,
and the structure and workings of the universe. One of the oldest was the
analogy between the sun, as the ruling power of the macrocosm, and the heart,
the governing power of the microcosm. Now at a medical school with the
reputation of that at Leyden, Hutton could not have taken the circulation of

2 1 trust that the significance of these words will be understood; no reader of Lowes
is likely to accuse Coleridge of plagiarism.

10 DONALD B. McINTYRE

blood as his subject without becoming very familiar with Harvey's classic
book "On the Movement of the Heart and Blood," published in 1628.

Harvey began his dedication, to King Charles I, thus: "The animal's heart
is the basis of its life, its chief member, the sun of its microcosm; on the heart
all its activity depends, from the heart all its liveliness and strength arise.
Equally is the king the basis of his kingdoms, the sun of his microcosm, the
heart of the state." And in his text Harvey wrote:

I began to think whether there might not be a motion, as it were in a circle,
in the same sense that Aristotle uses when he says that air and rain emulate
the circular movement of the heavenly bodies; for the moist earth warmed by
the sun evaporates; the vapors drawn upwards are condensed and fall as
rain to moisten the earth again, so producing successions of fresh life. In
similar fashion the circular movement of the sun gives rise to storms and
atmospheric phenomena. And so, in all likelihood, is it in the body, through
the motion of the blood. . . The heart deserves to be styled the starting
point of life and the sun of our microcosm just as much as the sun deserves
to be styled the heart of the world.

That the suggestion I am hinting at is not far-fetched is made clear by the
following quotation from Hutton: "The circulation of the blood is the efficient
cause of life; but life is the final cause, not only for the circulation of the blood,
but for the revolution of the globe: without a central luminary and a revolution
of the planetary body, there could not have been a living creature upon the
face of this earth." Now this quotation is not from his medical thesis, but from
the "Theory of the Earth," written 46 years after he left Leyden! And twice,
in that geologic classic, does Hutton refer to the "physiology" of the earth
a most significant phrase, and one reminiscent, incidentally, of Thomas
Robinson's book, "The Anatomy of the Earth," published in 1694, which pro-
claimed that the earth was a superorganism with "a constant circulation of
water, as in other animals of blood."

Analogy of microcosm and macrocosm, analogy of celestial spheres and
atmosphere, analogy of heart and sun, analogy of blood and rain: this is the
heredity of Hutton' s Theory of our theory. And the heart of the theory (if
I may use the analogy) is the concept of circulation of matter in the macrocosm.
One of the famous teachers at Leyden, just before Hutton' s time there, wrote
that "the author of nature has made it necessary for us to reason by analogy."
I know that Grove Karl Gilbert would have approved of this. The moral seems
to be that analogies are so important in the genesis of scientific hypotheses that
even false analogies are sometimes extremely fertile.

In closing I should like to report to you on the Circulation Club which was
founded in Edinburgh three years before Hutton's paper was read to the
Royal Society. Its object was, and here I quote from the Constitution: "to
commemorate the discovery of the circulation of the blood by the circulation

JAMES HUTTON AND THE PHILOSOPHY OF GEOLOGY 11

of the glass." Let us adopt and adapt that old tradition; let our hearts beat
faster and our blood thrill as we commemorate the discovery of circulation in
the macrocosm, and drink to the immortal name of the founder of geology,
James Hutton, M.D.

BIBLIOGRAPHY

BAILEY, Sir E. B., 1950, James Hutton, founder of modern geology (1726-1797): Roy.
Soc. Edinburgh, Pr., vol. 63, sec. B., 1947-49, pp. 357-368.

GILBERT, G. K., 1885, The inculcation of scientific method by example, with an illus-
tration drawn from the Quaternary geology of Utah: Am. J. Sci., 1886, 3rd series,
vol. 31, pp. 284-299.

, 1896, The origin of hypotheses, illustrated by the discussion of a topographic

problem: Science, n.s., vol. 3, pp. 1-13.

HARVEY, WILLIAM, 1628, Exercitatio anatomica de motu cordis et sanguinis in ani-
malibus: Tr. from original Latin by Kenneth J. Franklin for the Royal College of
Physicians of London, 1957, Oxford, Blackwell Scientific Publications, 209 pp.

HUTTON, JAMES, 1788, Theory of the earth; or an investigation of the laws observable
in the composition, dissolution, and restoration of land upon the globe: Roy. Soc.
Edinburgh, Tr., vol. 1, pt. 1, pp. 209-304.

, 1795, Theory of the earth, with proofs and illustrations: Edinburgh, Facsim.

reprint 2 vols., 1959, New York, Hafner.

LOWES, J. L., 1927, The road to Xanadu; a study in the ways of the imagination:
Boston, Houghton Mifflin, 623 pp.

PLAYFAIR, JOHN, 1802, Illustrations of the Huttonian theory of the earth: Edinburgh,
Cadell and Davies, and William Creech (Facsim. reprint with an introduction by
George W. White, 1956: Urbana, Univ. of Illinois Press), 528 pp.

, 1805, Biographical account of the late Dr. James Hutton, F.R.S. Edin.: Roy.

Soc. Edinburgh, Tr., vol. 5, pt. 3, pp. 39-99. Reprinted in The works of John
Playfair, Esq., 1822, vol. 4, pp. 33-118, Edinburgh.

POINCARE, HENRI, 1908, Science et m6thode: Translated by Francis Maitland, re-
printed 1952: New York, Dover Publications, 288 pp.

TOMKEIEFF, S. L, 1948, James Hutton and the philosophy of geology: Edinburgh
Geol. Soc., Tr., vol. 14, pt. 2, pp. 253-276. Reprinted 1950, Roy. Soc. Edinburgh,
Pr., vol. 63, sec. B, 1947-49, pp. 387-400.

, 1950, Geology in historical perspective: Adv. Sci., vol. 7, pp. 63-67.

TOULMIN, G. H., 1780, Antiquity and duration of the world: (cited by Tomkeieff).

, 1783, The antiquity of the world, 2d ed.: London, T. Cadell, 208 pp.

, 1785, The eternity of the world: (cited by Tomkeieff).

W. H. BRADLEY

U. S. Geological Survey

Geologic Laws 1

Students of the philosophy of science agree that general laws are numerous
and widely accepted among physicists and chemists. Indeed, they seem to
regard these laws as the sine qua non of science, and some even go so far as to
say that if there are no general laws, we are not dealing with true science but
with something of a lower order something more amorphous. Some philos-
ophers have also observed (though rarely in print) that, in geology, laws of
general scope are either rare or perhaps as yet unrecognized and therefore
unformulated.

General laws are indeed rare in geology. Why is this so? Is geology not
amenable to such generalizations or are geologists too little concerned with
the universal aspects of geology? Is geology perhaps still too immature to
produce generalizations of wide applicability or have geologists found that, for
their purposes, general laws are intellectual traps? Or do geologists, because
of their subject matter, have to reason somewhat differently from chemists
and physicists? Perhaps we, and the biologists who also must perforce travel a
comparable, or even more complex, network of paths, have something different
in the way of disciplined reasoning to offer the philosophers of science. Just
possibly, the philosophers of science err in judging the goals and caliber of a
science by the traditionally rigorous sciences of physics and chemistry. May
there not be goals other than the general laws of physics with their undeniably
beautiful simplicity and vast inclusiveness?

1 In writing this essay I have, as usual, had the benefit of generous criticism and
suggestions from a number of my colleagues in the U. S. Geological Survey. Of these,
I want to thank particularly W. T. Pecora and Dwight Taylor because, among other
things, they persuaded me to throw away the first draft and start afresh.

Besides this aid from my colleagues, I am especially grateful to Donald B. Mclntyre,
of Pomona College, and Allen M. Bassett, of San Diego State College, for constructive
suggestions. Professor Mclntyre also raised a number of provocative questions not
dealt with in my essay. Fortunately, most of his points are discussed and satisfyingly
answered by Simpson in his chapter.

12

GEOLOGIC LAWS 13

One wonders whether full understanding and satisfactory explanation of a
continuously operating dynamic process, such as the building and sculpture
of a mountain range or the division of living cells, are not as desirable goals
and as valuable rungs in the ladder of knowledge as the discovery of
Gay-Lussac's law. This is not to derogate Gay-Lussac's law; I merely wish to
raise for serious consideration a matter of long-term values in the realm of
knowledge. Actually, I wonder whether physicists, chemists, and indeed
scientists in general, are not today much less concerned with formulating laws
than they are with gaining deeper understanding of phenomena and the rela-
tionships among all kinds of phenomena.

It is pertinent to examine first what is meant by a law in science and then
take a brief look at geology itself. After that we shall be in a better position
to consider what laws there are in geology, what are some of the laws being
formulated today, and why laws are in fact rare in geology. But before we
get to these we shall consider briefly the more speculative aspects of the mental
tools geologists use in reasoning. These speculations involve an analogy with
modern biologists, whose problems of explication and law formulation seem
to have somewhat the same pattern as ours, and for similar reasons.

Of the many explanations of the term "law" in science, probably none is
more plainly stated than that of Karl Pearson (1900, pp. 86-87, 99):

Men study a range of facts . . . they classify and analyze, they discover re-
lationships and sequences, and then they describe in the simplest possible
terms the widest possible range of phenomena . . . We are thus to under-
stand by a law in science ... a resum6 in mental shorthand, which replaces
for us a lengthy description of the sequences among our sense impressions . . .
Such laws simply describe, they never explain . . .

Classical examples of natural laws are, of course, Newton's laws of motion
and the laws of thermodynamics. Some of the most fundamental relationships
in nature, such as the equivalence of mass and energy, are stated in the form
of general laws. They express concisely a group of more or less complex inter-
relationships that have repeatedly been observed to be consistent. They are
convenient packages of knowledge bench marks of a sort. They simplify our
efforts to explain the phenomena we study. But all such laws are man-made,
and for that reason reflect man's range of experience. They differ in degree of
generality. Some, like the laws of thermodynamics, are apparently valid for
all earthly phenomena over every yet conceivable range of values. Others are
valid for all ordinary conditions, say of pressure and temperature, but do not
hold strictly over extreme ranges, e.g., Boyle's law.

Natural laws that state more limited relationships are commonly referred to
as specific laws or empirical generalizations. They too are convenient small
bundles of knowledge. "Principle" is a word we, and other natural scientists,

14 W. H. BRADLEY

use freely, yet I have found no satisfactory way of discriminating between spe-
cific laws or empirical generalizations on the one hand and principles on the
other. Perhaps they are equivalents. If so, it is odd that geologists react so
cautiously toward one and so casually toward the other. A suitable illustration
of a "basic principle" is given by Gilluly, Waters, and Woodford (1951, p. 146)
in discussing the contribution Cuvier and Brongniart made to stratigraphy,
"Each formation (closely related group of strata) contains its own character-
istic assemblage of fossils."

Earlier in this essay I suggested that possibly geology has goals that are
somewhat different from those of physics and chemistry and has different
means for reaching them. If our ways of reasoning do in fact differ, the cause
must lie within the science of geology itself.

What are the properties of geology that set it apart? Geology is commonly
thought of as a derived science because it draws so much of its substance from
other sciences, notably chemistry, physics, and biology. But it has a hard core,
or spine, which is quite independent. It is this spine that gives geology its
distinctive qualities, and that determines in large measure the manner in
which its problems must be tackled. Geology's spine is the history and con-
stitution of the earth each term to be understood in its broadest sense. Run-
ning through this spine is the thread of time a thread connecting all terrestrial
events. The succession and relationships between these events, together with
the dynamic processes that have operated, make up the science of geology.
Geology's task is to reconstruct these events and explain their direct and in-
direct consequences. When we say "constitution of the earth" we mean to
include not only its elemental chemical composition but also its mineralogical
and lithologic constitution, the internal structures of minerals and rocks, and
the structural relationships these have with one another, i.e., the architecture
and ornamental features of the earth.

Long ago, Lyell's simple definition of geology (1832, p. 1) said much the
same thing:

Geology is the science which investigates the successive changes that have
taken place in the organic and inorganic kingdoms of nature; it inquires into
the causes of these changes, and the influence which they have exerted in
modifying the surface and external structure of our planet.

Indeed, everything we find in or on the earth came to be what it is, and where
it is, by geologic processes. Perhaps this should be qualified in the case of
organisms, but life also had its evolution in geologic history and certainly all
of life that has gone before us, and left any trace of itself or its activities, is of
concern to geologists.

To implement our understanding of geologic phenomena, we make use of
the dynamic earth processes of weathering, erosion, sedimentation, volcan-
ism and deformation all of which are assumed to have operated throughout

GEOLOGIC LAWS 15

the vast expanse of geologic time essentially as they do today. Nevertheless,
much of our understanding must come from completed "experiments" : from the
results of processes that required thousands or, more usually, millions of years
to form, or that have formed and lain dormant millions or billions of years.
The interpretation of such completed experiments is further complicated by
the fact that the same rocks may have been repeatedly deformed, and have
been reconstituted both chemically and mineralogically. Given immense
spans of time and immense forces acting for these long intervals, elements be-
come mobile and substances that are hard and brittle yield, flex, and flow.
As McKelvey points out elsewhere in this volume, most of these experiments
were of such a scale and involved forces and spans of time so enormous that
we probably cannot achieve adequate similitude in laboratory models designed
to test our explanations.

An additional difficulty is omnipresent, as only nongeologists need be re-
minded, and that is the impossibility of seeing all of most geologic features by
reason of their size and depth below the earth's surface, or by reason of a partial
cover of soil or younger rock, or by reason of the fact that parts of the features
have been eroded away. Therefore geologists are forever faced with the task
of reconstructing events that happened on a vast scale and in the remote past
from the partial remains of the products of those events. This compound
problem puts a premium on the capacity to reason analogically, inductively,
and with imagination.

Reasoning by analogy is not only common, it has probably been with us as
long as man has reasoned. Sir D'Arcy Thompson (1961, p. 6) refers to ana-
logical reasoning as another great Aristotelian theme. Indeed, he thus led into
the following footnote, "Hume declared, and Mill said much the same thing,
that all reasoning whatsoever depends on resemblance or analogy, and the
power to recognise it." G. K. Gilbert, apparently quite independently, ex-
pressed exactly the same concept to account for the way scientists originate
hypotheses (1896, pp. 2-3). I want to amplify Gilbert's specific point a little
and add a small homily for geologists. We can often say that the unknown
phenomenon resembles another phenomenon whose explanation is already
known or can be determined from experiment or from observation and study.
We are, in fact, often aware of present-day processes in nature which may
provide valuable analogies, only to find that the processes themselves are still
without adequate explanations. But these processes operate where we can
observe and measure them. Adequate explanations are within reach, if we will
but reach.

Because a geologist must use fragmentary data the incomplete results of
long-completed experiments he must use inductive reasoning to reconstruct
a whole from the parts. The virtue and value of inductive reasoning have
been extolled and used effectively by many. Others who have trod the same

16 W. H. BRADLEY

paths have been equally eloquent in pointing out the dangers inherent in the
use of inductive reasoning. Valuable and necessary as inductive reasoning is,
a geologist must keep at least one foot on the tangible earth which he seeks to
explain.

Many people equate imagination with the tendency to conjure up fanciful,
unreal things without restraint a faculty to be used either for amusement or
for escape from reality. And so it is, but in a more literal sense it is the power
of having mental images. Again, because a geologist can see only parts of the
features he studies and must forever deal with partial information (he constructs
geologic maps primarily to bring large features down to a comprehensible
scale at which he can integrate the parts and visualize the whole), it is most
essential that he be able to visualize, in three dimensions and with perspective,
processes that may have gone on that will help to reconstruct the events of the
past. Indeed, all the better if his imagination permits him to visualize processes
as they may have operated with time a sort of vision in motion. Although I
have mentioned imagination last of the three essential elements in the thought
processes of geologists, I am inclined to place it first in importance. A geologist
who has no imagination is as ineffective as a duck without webs between his
toes.

Perhaps such a long discursion on how geologists reason is out of place in
an essay on geologic laws. Nevertheless, by these mental means, I believe
geologists have made significant contributions. They have brought forth a
satisfying, though far from perfect, image of the earth on which we live and
a satisfying account of the long sequence of events that provided the changing
environments in which all life evolved. Such images and concepts have value
in human affairs and fill a significant niche in the realm of knowledge.

If this appraisal of the geologist's contribution is valid, it is pertinent to
enquire how much of this edifice was built on what kind of a foundation of
natural law. As W. M. Davis (1926, pp. 465-466) said, most of it is built on
inference the inference that the processes we observe today operated in the
same way in the remote past. But the events and the sequence of events are
none the less valid, for they have so far met all the tests we can apply to them
for the great expanse of time we can actually measure. For absolute measure-
ments of geologic time, however, we rely mostly on the physicists and their laws.

We might consider first how far geologists have gone and can go if they stay
within what we have defined as the core of geology that part which borrows
little or nothing from other disciplines. We can, with confidence, explain the
stratigraphic succession of any series of beds, and tell which way is "up" in
most such sequences, using purely geologic evidence. Indeed, the oft-repeated
verification that in any such sequence (not upset by tectonic movements) the
youngest beds are on top, gave rise many years ago to the law of superposition:
"In any pile of sedimentary strata that has not been disturbed by folding or

GEOLOGIC LAWS 17

overturning since accumulation, the youngest stratum is at the top and the
oldest at the base" (Gilluly, Waters, and Woodford, 1951, p. 73). This was
first clearly stated by Nicholas Steno in 1669. Steno also stated another law,
recognized by all geologists as generally valid, namely, the law of original
horizontality, thus: "Water-laid sediments are deposited in strata that are not
far from horizontal, and parallel or nearly parallel to the surface on which
they are accumulating" (idem, p. 73). We can tell the geologic ages of such
sequences from the fossils they contain, and we can correlate such sequences
of beds even though their outcrops are discontinuous. This follows the "basic
principle" established by Cuvier and Brongniart, already mentioned, and this
principle seems to have as good a claim to be one of our established laws as
those of Steno.

If we are granted the use of mathematics, we can, still within the framework
of pure geology, determine the sequence of events in complexly folded and
faulted areas, and can explain how we can predict where the various parts
are to be found beneath the surface. Furthermore, much of the sculpture of
the earth's surface can be explained without borrowing anything from other
sciences. (See, for example, Gilbert, 1877 and Hack, 1960.) Such explanations
must be based on analogy with present-day processes. Volcanic cones, ash
beds, lava flows, and mud flows all have ready explanations from what has
been observed in nature.

It was the analysis of a vast volume of such descriptions and explanations,
wholly within the realm of pure geology, that led Bucher (1933) to formulate
46 laws, each of which stated some generalization about the observed relation-
ships of deformational features of the earth's crust. Whether such generaliza-
tions (specific laws as we saw earlier in this essay) have served the purpose he
visualized, I leave to students of structural geology. Perhaps it will be re-
garded as another homily if I observe that another natural science, biology,
has formulated many such specific generalizations, or laws, and apparently has
used them to good advantage.

Except for Bucher's formulation of a goodly number of specific laws in 1933,
geologists of the present century have shown relatively little inclination to
formulate laws. But this was not always so. Seventy-five to 100 years ago
geologists were more prone to formulate generalizations and to state them as
laws. A few illustrations will suffice to make the point. In G. K. Gilbert's
Henry Mountains Report (1877, pp. 108-124) he analyzes the processes that
sculpture the land surface and speaks often about laws, although he explicitly
formulates only three: the law of declivity, "In general we may say that,
ceteris paribus, declivity bears an inverse relation to quantity of water"*; the law
of structure, "Insofar as the law of structure controls sculpture, hard masses
stand as eminences and soft are carved in valleys"; and the law of divides,
"The nearer the watershed or divide the steeper the slope."

18 W. H. BRADLEY

L. V. Pirsson (1905, p. 43) stated the following generalization as a "General
law of the province". "The petrographic province of central Montana is
characterized by the fact that in the most siliceous magmas the percentages
of potash and soda are about equal; with decreasing silica and increasing lime,
iron and magnesia, the potash relatively increases over the soda, until in the
least siliceous magmas it strongly dominates." Today such a generalization
would not be stated as a law. This example, and those cited above from
Gilbert, reflect a tendency of those times a striving for generalization and for
fixing generalizations in the form of laws. From our present vantage point,
realizing now how complex nature actually is, we probably would be right in
saying that such a tendency reflected a then unrecognized immaturity of our
science.

If geology was too immature 75 or 100 years ago to warrant the formulation
of general laws, what is our status today?

Actually, if one were to make an exhaustive census, which I have not, we
probably would find more generalizations being expressed now than there
were a generation or two ago. Three illustrations with which I happen to be
acquainted will serve to characterize the current effort, though none is called
or thought of as a law.

One is the principle of dynamic equilibrium in the evolution of landscape
elaborated by Hack (1960, pp. 85-96) from earlier statements by G. K. Gilbert,
W. M. Davis, and A. N. Strahler. According to Hack (p. 86):

The concept requires a state of balance between opposing forces such that
they operate at equal rates and their effects cancel each other to produce a
steady state, in which energy is continually entering and leaving the system.
The opposing forces might be of various kinds. For example, an alluvial
fan would be in dynamic equilibrium if the debris shed from the mountain
behind it were deposited on the fan at exactly the same rate as it was removed
by erosion from the surface of the fan itself.

Another generalization was expressed by T. B. Nolan in his presidential ad-
dress before the Geological Society of America (1962, p. 277). In discussing
the grade of mineral resources he pointed out that in making estimates of future
resources one should keep in mind not only the fact that we have been able
over the years to use progressively lower-grade ores, but "that the tonnages of
material available in the earth's crust increase geometrically with decreasing
grade."

Still another generalization or, more accurately, a group of generalizations,
was presented by James B. Thompson (1955, pp. 65-103) in his treatment of
the thermodynamic basis for the mineral facies concept.

He considers the following four limiting cases and then discusses certain
geologic applications.

GEOLOGIC LAWS 19

CASE 1 . The mobile component in question is present as a pure phase at
the same temperature and pressure as the surrounding rock. This is the
assumption implicit in current hydrothermal experimental work and in
calculated equilibria such as that of Goldschmidt for calcite-wollaston-
ite. It yields a maximum equilibrium temperature at any given pressure
(depth).

CASE 2. The rock is in chemical equilibrium \\ ith a fissure system containing
a dilute aqueous solution. This method is applicable only when the mobile
component in question is water itself. It indicates a decrease in most de-
hydration temperatures with depth.

CASE 3. The vertical gradient of chemical potential of the mobile compo-
nent has reached a value such that there is no longer any tendency for
vertical diffusion of this component. This method also suggests (not without
reservations) a slight decrease of most devolatilization temperatures with
depth.

CASE 4. The chemical potential of the mobile component is sufficiently low
that no phases containing it are stable.

For simplicity, we will refer to these limiting cases as cases 1, 2, 3, and 4,
respectively. It now remains for us to consider the extent to which these
may be applicable to some specific geological problems, (p. 96)

Thompson's treatment of the various mineralogical changes under the as-
sumed conditions goes far toward quantifying these geologic processes and
therefore constitutes an important advance in the science of geology. Given
adequate data on the environments and thermochemical data on the minerals
involved, it appears that similar generalizations can now be made about such
processes as diagenesis, weathering, and the formation of ore deposits.

Thus far (with the exception of Thompson's work just mentioned) we have
deliberately kept within that sphere of geology which is purely geologic, that
is, not borrowed from any other science. We have seen that we can explain
many of the phenomena of geology within this frame of reference. But such
explanations leave unanswered virtually all questions of "how." How can
hard rocks bend and flow; what caused this mineral to change over into that,
what permits it to do so and what are the conditions that determine the change;
how do we account for the fact that the ratios of oxygen isotopes change pro-
gressively with distance from a certain focus; how can such things as amino
acids be preserved in fossils for millions of years; what must have been the
conditions to permit the preservation of the color pattern in a fossil or a chloro-
plast within a plant cell? How can we account for the mobility and rearrange-
ment of elements and compounds within the earth's crust; how does it come
about that certain elements aggregate to form deposits of commercial value

20 W. H. BRADLEY

and that some minerals are piezoelectric? One needs only to begin asking
"how" in geology and the list grows like weeds after a warm rain.

We cannot begin to answer these queries without the help of chemistry,
physics, or biology. In every problem of this broad class of problems we realize
that minerals, the fundamental units of geology, are chemical compounds and
that they are what they are and came to be where they are by chemical as well
as geologic processes. Moreover, they have optical, electrical, magnetic, ther-
mal, and mechanical properties, and atomic structures that belong just as much
in the realm of physics as the properties of any man-made material or man-
designed experiment. In short, we can scarcely look to a deeper understanding
of any of our geologic problems without coming face to face with the realiza-
tion that all the materials we seek to understand and whose behavior we seek
to understand have the properties, structures, and behavior of all matter. The
answers we seek must come from the same kinds of sources that the chemist,
the physicist, and the biologist study to get their enlightenment.

We are utterly dependent on chemistry, physics, and biology for under-
standing, yet the problems we bring up from the earth do not in any sense
thereby lose their geologic character. Geologists must and will continue to
observe, measure, and analyze geologic phenomena and, more importantly,
they will continue to formulate the problems to ask exactly how and exactly
why. This is the province and the responsibility of geologists, and no one can
be expected to do it better. Important as chemistry and physics are to the
solution of geologic problems, neither chemists nor physicists are going to take
over the role of formulating the problems that arise out of geologic work unless
they themselves become geologists. This is not to say, of course, that the geo-
chemists and geophysicists will not generate questions of their own about earth
processes.

We have considered what natural laws are, what properties give geology its
distinguishing marks, something about the way geologists reason, and lastly
the growing realization of the great complexity of geology and the growing
dependence of geology on the disciplines of chemistry, physics, and biology.
Let us consider next the steps in the development of generalizations in geology.
At least in part this can be expressed graphically (Fig. 1). The concept of
such a graph evolved in a discussion with my colleague V. E. McKelvey, after
he had returned from the Brookings Institution's Conference for Federal
Science Executives at Williamsburg, Va., where he had heard Henry Margenau
of Yale discuss the development of natural laws in general. Margenau's
diagram, as I understand it, consisted of three ordinates, labeled Primary
observations, Concepts, and Propositions (laws). These are used also in Fig. 1.
McKelvey and I added the boundary curve to indicate the philosophical field
of activity of geologists. The great bulk of geologic activity hovers close to the
observational ordinate. When enough observations have been related and
explained, the activity moves right into the field of concepts. This move comes

GEOLOGIC LAWS

21

Primary
observations

Concepts
(principles)

Propositions
(laws)

Generalization *-

Fig. 1. Diagrammatic representation of the field of geologic activity and thought
plotted against Margenau's three ordinates: Primary observations, Concepts, and
Propositions (laws). The shaded field above represents a reservoir of the combined
disciplines of chemistry, physics, and biology in an unsorted mixture the position of
the names, therefore, has no relationship to Margenau's ordinates. The crosshatched
part of this field indicates progressively greater assimilation by geology of chemistry,
physics, and biology. Ideas in geology move out along the generalization abscissa in
proportion to the number and significance of the verified observations and theories
that are unified into broader concepts or principles.

about from inductive reasoning and results in a concept or principle, i.e., a
generalization, which we repeatedly test against more observations. Paren-
thetically, we also use these principles to explicate other geologic phenomena,
thereby nourishing a sort of feedback mechanism, which is present in the
reasoning operations of all sciences. But I am inclined to think that such
feedback operations are especially important in geology because we deal so
much with fragmentary data and with working hypotheses.

The theory of isostasy, the origin of igneous rocks, and organic evolution
are examples of such concepts or principles.

Most geologists are happy to explore and describe. Fewer are unhappy until
they are able also to explain what they have found. And very few seem in-
terested in trying to integrate the explanations into more general concepts or
principles. Just possibly this derives from the patent fact that it is a lot more
fun to do field work than it is to put the pieces of explained phenomena to-
gether, especially if the pieces were constructed by someone else.

Casual inspection of this curve (Fig. 1) prompts one to ask why (with the
exception of Steno's two laws formulated long ago) we have not, or do not,
push our concepts or principles still farther to the right into the realm of laws?
I believe there are several reasons for this. One is the growing realization that

22 W. H. BRADLEY

the earth's history and the processes involved in shaping it are vastly more
complex than was thought even a generation or tvvo ago. We have, so to speak,
become more conscious of the number of variables and the magnitude of the
gaps in our knowledge. Another reason is the growing realization that we can
press our effort for deeper understanding of geology only with the aid of pro-
gressively more chemistry, physics, and biology. As we progress with our
reasoning toward the right in the diagram, the content of these borrowed
disciplines must increase. I hazard the guess that the area under the curve in
the region of the "concept" and "laws" ordinates will increase with time in
proportion to geology's assimilation of chemistry, physics, and biology.

In conclusion, it seems to me that wherever we look there is a striving for
deeper understanding of ever more complex problems and a corresponding
realization of how much more we must learn, especially in geochemistry and
geophysics, in order to gain this fuller understanding. Such phases of rapid
expansion of the frontiers of any science, I suggest, are not conducive to think-
ing about the formulation of general laws. Everything is too new, too pregnant
with the possibility of growth and change too much is clearly missing or
imperfectly understood for us to become preoccupied with reflective thinking
about that which is already known. The new and unsolved problems are much
more worth the candle, are more exciting, more rewarding. Reflection is
crowded out in the excitement of the day.

The biologists appear to find themselves in pretty much the same situation.
Paul Weiss (1962, p. 470) stated it thus:

Biology ... is destined to retain its autonomy, which means that to be
known and understood, biological mechanisms will have to be studied and
formulated in their own right and full diversity. And this explains why in
biology so many generalizations must stop far short of the vast inclusiveness
of laws of physics, hence, why the range of validity of each must be deter-
mined empirically. It is this inherent feature of biological nature, rather
than backwardness or extravagance, then, which necessitates testing over a
far wider spectrum of variables . . . than would seem necessary or even
pardonable in most of physics.

But this situation we find ourselves in is no cause for lament; rather is it a
source of inspiration, a cause of exhilaration. It seems to me geology is now
at the beginning of an explosive stage of development comparable to that which
occurred in the biological sciences when biochemistry and biophysics came into
full swing. Let the philosophers of science revel a bit in the yeasty ferment that
is geology today and tomorrow. Some day we may grow old and have more
laws; right now we are busy exploring, experimenting, and trying to understand
more of the "how" of those processes that have produced the features of the
earth, its crust beneath, and all it contains.

GEOLOGIC LAWS 23

REFERENCES CITED

BUCHER, W. H., 1933, The deformation of the earth's crust: Princeton, Princeton Univ

Press, 518 pp.

, 1936, The concept of natural law in geology: Ohio J. Sci., vol. 36, pp. 183-194.

DAVIS, W. M., 1926, The value of outrageous geological hypotheses: Science, vol. 63,

pp. 463-468.
GILBERT, G. K., 1877, Geology of the Henry Mountains: U. S. Geog. and Geol. Survey

of the Rocky Mountain Region, Washington, D. C., Government Printing Office,

170 pp.
, 1896, The origin of hypotheses, illustrated by the discussion of a topographic

problem: Science, n.s., vol. 3, pp. 1-13.
GILLULY, J., WATERS, A. C., and WOODFORD, A. O., 1951, Principles of geology:

San Francisco, W. H. Freeman, 631 pp.
HACK, J. T., 1960, Interpretation of erosional topography in humid temperate regions:

Am. J. Sci., vol. 258-A, pp. 80-97.

LYELL, Sir CHARLES, 1832, Principles of geology, 2nd ed.: London, John Murray, vol. 1.
NOLAN, T. B., 1962, Role of the geologist in the national economy: Geol. Soc. Am., B.,

vol. 73, pp. 273-278.
PEARSON, KARL, 1900, The grammar of science, 2d ed.: London, Adam and Charles

Black, 548 pp.
PIRSSON, L. V., 1905, The petrographic province of central Montana: Am. J. Sci.,

vol. 20 (whole no. 170), pp. 35-49.
THOMPSON, Sir D'ARCY W., 1961, On growth and form (abridged edition, edited by

J. T. Bonner): New York, Cambridge Univ. Press, 346 pp.
THOMPSON, J. B., 1955, The thermodynamic basis for the mineral facies concept:

Am. J. Sci., vol. 253, pp. 65-103.
WEISS, PAUL, 1962, Experience and experiment hi biology: Science, vol. 136, pp. 468-

471.

GEORGE GAYLORD SIMPSON

Harvard University

Historical Science 1

History, Science, and Historical Science

The simplest definition of history is that it is change through time. It is,
however, at once clear that the definition fails to make distinctions which are
necessary if history is to be studied in a meaningful way. A chemical reaction
involves change through time, but obviously it is not historical in the same
sense as the first performance by Lavoisier of a certain chemical experiment.
The latter was a nonrecurrent event, dependent on or caused by antecedent
events in the life of Lavoisier and the lives of his predecessors, and itself causal
of later activities by Lavoisier and his successors. The chemical reaction
involved has no such causal relationship and has undergone no change before
or after Lavoisier's experiment. It always has occurred and always will recur
under the appropriate historical circumstances, but as a reaction in itself it has
no history.

A similar contrast between the historical and the nonhistorical exists in
geology and other sciences. The processes of weathering and erosion are un-
changing and nonhistorical. The Grand Canyon or any gully is unique at
any one time but is constantly changing to other unique, nonrecurrent con-
figurations as time passes. Such changing, individual geological phenomena
are historical, whereas the properties and processes producing the changes
are not.

The unchanging properties of matter and energy and the likewise unchanging
processes and principles arising therefrom are immanent in the material universe.
They are nonhistorical, even though they occur and act in the course of history.
The actual state of the universe or of any part of it at a given time, its con-
figuration, is not immanent and is constantly changing. It is contingent in

1 Parts of this essay have been developed from a talk on the explanation of unique
events given in the Seminar on Methods in Philosophy and Science at the New School
for Social Research on May 20, 1962. On that occasion I also profited from discussion
and other talks pertinent to the present topic, especially by Dobzhansky, Nagel, and
Pittendrigh.

24

HISTORICAL SCIENCE 25

BernaTs (1951) term, or configurational, as I prefer to say (Simpson, 1960).
History may be defined as configurational change through time, i.e., a sequence
of real, individual but interrelated events. These distinctions between the
immanent and the configurational and between the nonhistorical and the
historical are essential to clear analysis and comprehension of history and of
science. They will be maintained, amplified, and exemplified in what follows.

Definitions of science have been proposed and debated in innumerable ar-
ticles and books. Brief definitions are inevitably inadequate, but I shall here
state the one I prefer: Science is an exploration of the material universe that
seeks natural, orderly relationships among observed phenomena and that is
self-testing. (For explanation and amplification see Simpson, 1962, 1963.)
Apart from the points that science is concerned only with the material or
natural and that it rests on observation, the definition involves three scientific
activities: the description of phenomena, the seeking of theoretical, explanatory
relationships among them, and some means for the establishment of confidence
regarding observations and theories. Among other things, later sections of
this essay will consider these three aspects of historical science.

Historical science may thus be defined as the determination of configurational
sequences, their explanation, and the testing of such sequences and explana-
tions. (It is already obvious and will become more so that none of the three
phases is simple or thus sufficiently described.)

Geology is probably the most diverse of all the sciences, and its status as in
part a historical science is correspondingly complex. For one thing, it deals
with the immanent properties and processes of the physical earth and its con-
stituents. This aspect of geology is basically nonhistorical. It can be viewed
simply as a branch of physics (including mechanics) and chemistry, applying
those sciences to a single (but how complex!) object: the earth. Geology also
deals with the present configuration of the earth and all its parts, from core
to atmosphere. This aspect of geology might be considered nonhistorical
insofar as it is purely descriptive, but then it also fails to fulfill the whole
definition of a science. As soon as theoretical, explanatory relationships are
brought in, so necessarily are changes and sequences of configurations, which
are historical. The fully scientific study of geological configurations is thus
historical science. This is the only aspect of geology that is peculiar to this
science, that is simply geology and not also something else. (Of course I do
not mean that it can be studied without reference to other aspects of geology
and to other sciences, both historical and nonhistorical.)

Paleontology is primarily a historical science, and it is simultaneously bio-
logical and geological. Its role as a part of historical biology is obvious. In
this role, like all other aspects of biology, it involves all the immanent properties
and processes of the physical sciences, but differs from them not only in being
historical but also in that its configurational systems are incomparably more

26 GEORGE GAYLORD SIMPSON

complex and have feedback and information storage and transmittal mechan-
isms unlike any found in the inorganic realm. Its involvement in geology and
inclusion in that science as well as in biology are primarily due to the fact that
the history of organisms runs parallel with, is environmentally contained in,
and continuously interacts with the physical history of the earth. It is of less
philosophical interest but of major operational importance that paleontology-,
when applicable, has the highest resolving power of any method yet dis-
covered for determining the sequence of strictly geological events. (That
radiometric methods may give equal or greater resolution is at present a hope
and not a fact.)

Description and Generalization

In principle, the observational basis of any science is a straight description
of what is there and what occurs, what Lloyd Morgan (1891) used to call
"plain story." In a physical example, plain story might be the specifications
of a pendulum and observations of its period. A geological plain story might
describe a bed of arkose, its thickness, its attitude, and its stratigraphic and
geographic position. An example of paleontological plain story would be the
occurrence of a specimen of a certain species at a particular point in the bed
of arkose. In general, the more extended plain stories of historical science
would describe configurations and place them in time.

In fact, plain story in the strictest, most literal sense plays little part in science.
Some degree of abstraction, generalization, and theorization usually enters in,
even at the first observational level. The physicist has already abstracted a
class of configurational systems called "pendulums" and assumes that only the
length and period need be observed, regardless of other differences in indi-
viduals of the class, unless an observation happens to disagree with the assump-
tion. Similarly, the geologist by no means describes all the characteristics of
the individual bed of arkose and its parts but has already generalized a class
"arkose" and adds other details, if any, only in terms of such variations within
the class as are considered pertinent to his always limited purpose. The
paleontologist has departed still further from true, strict plain story, for in
recording a specimen as of a certain species he has not only generalized a
particularly complex kind of class but has also reached a conclusion as to
membership in that class that is not a matter of direct observation at all.

Every object and every event is unique if its configurational aspects are
described in full. Yet, and despite the schoolteachers, it may be said that some
things are more unique than others. This depends in the first place on the com-
plexity of what is being described, for certainly the more complex it is the more
ways there are in which it may differ from others of its general class. A bed
of arkose is more complex than a pendulum, and an organism is to still greater

HISTORICAL SCIENCE 27

degree more complex than a bed of arkose. The hierarchy of complexity and
individual uniqueness from physics to geology to biology is characteristic of
those sciences and essential to philosophical understanding of them. It bears
on the degree and kind of generalization characteristic of and appropriate to
the various sciences even at the primary observational level. The number of
pertinent classes of observations distinguished in physics is much smaller than
in geology, and much smaller in geology than in biology. For instance, in
terms of taxonomically distinguished discrete objects, compare the numbers of
species of particles and atoms in physics, of minerals and rocks in geology, and
of organisms in biology. Systems and processes in these sciences have the same
sequence as to number and complexity.

Another aspect of generalization and degree of uniqueness arises in com-
parison of nonhistorical and historical science and in the contrast between
immanence and configuration. In the previous examples, the physicist was
concerned with a nonhistorical and immanent phenomenon: gravitation. It
was necessary to his purpose and inherent in his method to eliminate as far as
possible and then to ignore any historical element and any configurational
uniqueness in the particular, individual pendulum used in the experiment.
He sought a changeless law that would apply to all pendulums and ultimately
to all matter, regardless of time and place. The geologist and paleontologist
were also interested in generalization of common properties and relationships
between one occurrence of arkose and another, between one specimen and
another of a fossil species, but their generalizations were of the configurational
and not the immanent properties and were, or at least involved, historical and
not only nonhistorical science. The arkose or the fossil had its particular as
well as its general configurational properties, its significant balance of difference
and resemblance, not only because of immanent properties of its constituents
and immanent processes that had acted on it, but also because of its history,
the configurational sequence by which these individual things arose. The latter
aspect, not pertinent to the old pendulum experiment or to almost anything in
the more sophisticated physics of the present day, is what primarily concerns
geology and paleontology' as historical sciences, or historical science in general.

Scientific Law

It has been mentioned that the purpose of the pendulum experiment was to
formulate a law. The concept of scientific law and its relationship with his-
torical and nonhistorical science are disputed questions requiring clarification.
The term "law" has been so variously and loosely used in science that it is no
longer clear unless given an explicit and restrictive definition. The college
dictionary that I happen to have at hand (Barnhart, 1948) defines "law" in
philosophical or scientific use as "a statement of a relation or sequence of

28 GEORGE GAYLORD SIMPSON

phenomena invariable under the same conditions." This is satisfactory if it is
made clear that a law applies to phenomena that are themselves variable: it
is the relationship (or sequence, also a relationship) that is invariable. "Under
the same conditions" must be taken to mean that other variables, if present,
are in addition to and not inextricably involved with those specified in the law.
Further, it is perhaps implicit but should be explicit that the relationship must
be manifested or repeatable in an indefinitely recurrent way. A relationship
that could or did occur only once would indeed be invariable, but surely would
not be a law in any meaningful scientific sense. With these considerations, the
definition might be rephrased thus: a scientific law is a recurrent, repeatable
relationship between variables that is itself invariable to the extent that the
factors affecting the relationship are explicit in the law.

The definition implies that a valid law includes all the factors that necessarily
act in conjunction. The fact that air friction also significantly affects the ac-
celeration of a body falling in the atmosphere does not invalidate the law of
gravitational acceleration, but only shows that the body is separately acted on
by some factor defined by another law. Friction and the factors of gravitational
acceleration are independent. Both laws are valid, and they can be combined
into a valid compound law. But if some factor necessarily involved in either one,
such as force of gravitation for acceleration or area for friction, were omitted,
that law would be invalidated.

Laws, as thus defined, are generalizations, but they are generalizations of a
very special kind. They are complete abstractions from the individual case.
They are not even concerned with what individual cases have in common, in
the form of descriptive generalizations or definitions, such as that all pendulums
are bodies movably suspended from a fixed point, all arkoses are sedimentary
rocks containing feldspar, or all vertebrates are animals with jointed backbones.
These and similar generalizations are obviously not laws by any usage. When
we say, for instance, that arkose is a feldspathic sedimentary rock, we mean
merely that we have agreed that if a rock happens to be sedimentary and
within a certain range of texture and of composition including a feldspar, we
will call it "arkose." We do not mean that the nature of the universe is such
that there is an inherent relationship among sedimentary rocks and feldspars
reducible to a constant. Laws are inherent, that is immanent, in the nature of
things as abstracted entirely from contingent configurations, although always
acting on those configurations.

Until recently the theoretical structure of the nonhistorical physical sciences
consisted largely of a body of laws or supposed laws of this kind. The prestige
of these sciences and their success in discovering such laws were such that it
was commonly believed that the proper scientific goal of the historical sciences
was also to discover laws. Supposed laws were proposed in all the historical
sciences. By way of example in my own field, paleontology, I may mention

HISTORICAL SCIENCE 29

"Dollo's law" that evolution is irreversible, "Cope's law" that animals become
larger in the course of evolution, or "Williston's law" that repetitive serial
structures in animals evolve so as to become less numerous but more differen-
tiated. The majority of such supposed laws are no more than descriptive gen-
eralizations. For example, animals do not invariably become larger in time.
Cope's law merely generalizes the observation that this is a frequent tendency,
without establishing any fixed relationship among the variables possibly in-
volved in this process.

Even when a relationship seems established, so-called "historical laws" are
almost always open to exceptions. For example, Rensch (1960), an evolutionist
convinced of the validity of historical laws, considers "Allen's rule" a law:
that when mammals adapt to colder climates their feet become shorter. But
of the actual mammals studied by him, 36% were exceptions to the "law."
Rensch explains this by supposing that "many special laws act together or
interfere with one another. Thus 'exceptions' to the laws result." This is a
hypothetical possibility, but to rely upon it is an act of faith. The "inter-
fering" laws are unknown in this or similar examples. A second possibility is
that the "laws," as stated, are invalid as laws because they have omitted factors
necessarily and inherently involved. I believe this is true, not in the sense that
we have only to complete the analysis and derive a complete and valid law,
but in the sense that the omissions are such as to invalidate the very concept
of historical law.

The search for historical laws is, I maintain, mistaken in principle. Laws
apply, in the dictionary definition "under the same conditions," or in my
amendment "to the extent that factors affecting the relationship are explicit in
the law," or in common parlance "other things being equal." But in history,
which is a sequence of real, individual events, other things never are equal.
Historical events, whether in the history of the earth, the history of life, or
recorded human history, are determined by the immanent characteristics of
the universe acting on and within particular configurations, and never by either
the immanent or the configurational alone. It is a law that states the relation-
ship between the length of a pendulum any pendulum and its period. Such
a law does not include the contingent circumstances, the configuration, neces-
sary for the occurrence of a real event, say Galileo's observing the period of a
particular pendulum. If laws thus exclude factors inextricably and significantly
involved in real events, they cannot belong to historical science.

It is further true that historical events are unique, usually to a high degree,
and hence cannot embody laws defined as recurrent, repeatable relationships.
Apparent repetition of simple events may seem to belie this. A certain person's
repeatedly picking up and dropping a certain stone may seem to be a recurrent
event in all essentials, but there really is no applicable historical law. Abstrac-
tion of a law from such repeated events leads to a nonhistorical law of immanent

30 GEORGE GAYLORD SIMPSON

relationships, perhaps in this case of gravity and acceleration or perhaps of
neurophysiology, and not to a historical law of which this particular person,
picking up a certain stone, at a stated moment, and dropping it a definite
number of times would be a determinate instance. In less trivial and more
complex events, it is evident that the extremely intricate configurations in-
volved in and necessary, for example, as antecedents for the erosion of the
Grand Canyon or the origin of Homo sapiens simply cannot recur and that there
can be no laws of such one-of-a-kind events. (Please bear in mind that the true,
immanent laws are equally necessary and involved in such events but that they
remain nonhistorical; the laws would have acted differently and the historical
event, the change of configuration, would have been different if the configura-
tion had been different; this historical element is not included in the operative
laws.)

It might be maintained that my definition of law is old-fashioned and is no
longer accepted in the nonhistorical sciences either. Many laws of physics,
considered nonhistorical, are now conceived as statistical in nature, involving
not an invariable relationship but an average one. The old gas laws or the
new laws of radioactive decay are examples. The gas laws used to assume an
ideal gas. Now they are recognized as assuming that directions of molecular
motion tend to cancel out if added together, and that velocities tend to vary-
about a mean under given conditions. This cannot be precisely true of a real
gas at a given moment, but when very large numbers of molecules are involved
over an appreciable period of time, the statistical result is so close to the state
described by the gas laws that the difference does not matter. In this and similar
ways the descent from the ideal to the real in physical science has been coped
with, not so much by facing it as by finding devices for ignoring it.

The historical scientist here notes that a real gas in a real experiment has
historical attributes that are additional to the laws affecting it. Every molecule
of a real gas has its individual history. Its position, direction of motion, and
velocity at a given moment (all parts of the total configuration) are the outcome
of that history. It is, however, quite impractical and, for the purposes of
physics, unnecessary to make an historical study of the gas. The gas laws apply
well enough "other things being equal," which means here that the simple
histories of the molecules tend, as observation shows, to produce a statistical
result so nearly uniform that the historical, lawless element can be ignored for
practical purposes.

The laws immanent in the material universe are not statistical in essence.
They act invariably in variable historical circumstances. The pertinence of
statistics to such laws as those of gases is that they provide a generalized de-
scription of usual historical circumstances in which those laws act, and not
that they are inherent in the laws themselves. Use of statistical expressions,
not as laws but as generalized descriptions, is common and helpful in all science

HISTORICAL SCIENCE 31

and especially in historical science. For example, the statistical specifications
of land forms or of grain size in sediments clearly are not laws but descriptions
of configurations involved in and arising from history.

To speak of "laws of history" is either to misunderstand the nature of history
or to use "laws" in an unacceptable sense, usually for generalized descriptions
rather than formulations of immanent relationships.

Uniform! tarianism

Uniformitarianism has long been considered a basic principle of historical
science and a major contribution of geology to science and philosophy. In one
form or another it does permeate geological and historical thought to such a
point as often to be taken for granted. Among those who have recently given
conscious attention to it, great confusion has arisen from conflicts and ob-
scurities as to just what the concept is. To some, uniformitarianism (variously
defined) is a law of history. Others, maintaining that it is not a law, have
tended to deny its significance. Indeed, in any reasonable or usual formulation,
it is not a law, but that does not deprive it of importance. It is commonly
defined as the principle that the present is the key to the past. That definition
is, however, so loose as to be virtually meaningless in application. A new,
sharper, and clearer definition in modern terms is needed.

Uniformitarianism arose around the turn of the 18th to 19th centuries, and
its original significance can be understood only in that context. (The historical
background is well covered in Gillispie, 1951.) It was a reaction against the
then prevailing school of catastrophism, which had two main tenets: (1) the
general belief that God has intervened in history, which therefore has included
both natural and supernatural (miraculous) events; and (2) the particular
proposition that earth history consists in the main of a sequence of major
catastrophes, usually considered as of divine origin in accordance with the
first tenet. (For a historical review anachronistically sympathetic with these
beliefs see Hooykaas, 1959.) Uniformitarianism, as then expressed, had various
different aspects and did not always face these issues separately and clearly.
On the whole, however, it embodied two propositions contradictory to catas-
trophism: (1) earth history (if not history in general) can be explained in terms
of natural forces still observable as acting today; and (2) earth history has not
been a series of universal or quasi-universal catastrophes but has in the main
been a long, gradual development what we would now call an evolution.
(The term "evolution" was not then customarily used in this sense.) A classic
example of the conflicting application of these principles is the catastrophist
belief that valleys are clefts suddenly opened by a supernally ordered revolution
as against the uniformitarian belief that they have been gradually formed by
rivers that are still eroding the valley bottoms.

32 GEORGE GAYLORD SIMPSON

Both of the major points originally at issue are still being argued on the
fringes of science or outside it. To most geologists, however, they no longer
merit attention from anyone but a student of human history. It is a necessary
condition and indeed part of the definition of science in the modern sense that
only natural explanations of material phenomena are to be sought or can be
considered scientifically tenable. It is interesting and significant that general
acceptance of this principle (or limitation, if you like) came much later in the
historical than in the nonhistorical sciences. In historical geology it was the
most important outcome of the uniformitarian-catastrophist controversy. In
historical biology it was the still later outcome of the Darwinian controversy
and was hardly settled until our own day. (It is still far from settled among
nonscientisls.)

As to the second major point originally involved in uniformitarianism, there
is no a priori or philosophical reason for ruling out a series of natural worldwide
catastrophes as dominating earth history. However, this assumption is simply
in such flat disagreement with everything we now know of geological history
as to be completely incredible. The only issues still valid involve the way in
which natural processes still observable have acted in the past and the sense
in which the present is a key to the past. Uniformitarianism, or neo-uni-
formitarianism, as applied to these issues has taken many forms, among them
two extremes that are both demonstrably invalid. They happen to be rather
amusingly illustrated in a recently published exchange of letters by Lippman
(1962) andFarrand (1962).

Lippman, one of the neocatastrophists still vociferous on the fringes of
geological science, attacks uniformitarianism on the assumption that its now
"orthodox" form is absolute gradualism, i.e., the belief that geological processes
have always acted gradually and that changes catastrophic in rate and extent
have never occurred. Farrand, who would perhaps consent to being called an
orthodox geologist, demonstrates that Lippman has set up a straw man.
Catastrophes do now occur. Their occurrence in the past exemplifies rather
than contradicts a principle of uniformity. It happens that there is no valid
evidence that catastrophes of the kind and extent claimed by the original
catastrophists and by Lippman have ever occurred or that they could provide
explanations for some real phenomena, as claimed. This, however, is a different
point. Farrand expresses a common, probably the usual modern understanding
of uniformitarianism as follows: "The geologist's concept that processes that
acted on the earth in the past are the same processes that are operating today,
on the same scale and at approximately the same rates" (italics mine). But this
principle also seems to be flatly contradicted by geological history. Some pro-
cesses (those of vulcanism or glaciation, for example) have evidently acted in
the past on scales and at rates that cannot by any stretch be called "the same"
or even "approximately the same" as those of today. Some past processes

HISTORICAL SCIENCE 33

(such as those of Alpine nappe formation) are apparently not acting today, at
least not in the form in which they did act. There are innumerable exceptions
that disprove the rule.

Then what uniformity principle, if any, is valid and important? The dis-
tinction between immanence and configuration (or contingency) clearly points
to one: the postulate that immanent characteristics of the material universe
have not changed in the course of time. By this postulate all the immanent
characteristics exist today and so can, in principle, be observed or, more
precisely, inferred as generalizations and laws from observations. It is in this
sense that the present is the key to the past. Present immanent properties and
relationships permit the interpretation and explanation of history precisely
because they are not historical. They have remained unchanged, and it is the
configurations that have changed. Past configurations were never quite the
same as they are now and were often quite different. Within those different
configurations, the immanent characteristics have worked at different scales
and rates at different times, sometimes combining into complex processes
different from those in action today. The uniformity of the immanent charac-
teristics helps to explain the fact that history is not uniform. (It could even be
said that uniformitarianism entails catastrophes, but the paradox would be
misleading if taken out of context.) Only to the extent that past configurations
resembled the present in essential features can past processes have worked in a
similar way.

That immanent characteristics are unchanging may seem at first sight either
a matter of definition or an obvious conclusion, but it is neither. Gravity would
be immanent (an inherent characteristic of matter now) even if the law of
gravity had changed, and it is impossible to prove that it has not changed.
Uniformity, in this sense, is an unprovable postulate justified, or indeed re-
quired, on two grounds. First, nothing in our incomplete but extensive knowl-
edge of history disagrees with it. Second, only with this postulate is a rational
interpretation of history possible, and we are justified in seeking as scientists
we must seek such a rational interpretation. It is on this basis that I have
assumed on previous pages that the immanent is unchanging.

Explanation

Explanation is an answer to the question "Why?" But as Nagel (1961) has
shown at length, this is an ambiguous question calling for fundamentally dif-
ferent kinds of answers in various contexts. One kind of answer specifies the
inherent necessity of a proposition, and those are the answers embodied in laws.
Some philosophers insist that this is the only legitimate form of explanation.
Some (e.g. Hobson, 1923) even go so far as to maintain that since inherent
necessity cannot be proved, there is no such thing as scientific explanation.

34 GEORGE GAYLORD SIMPSON

Nagel demonstrates that all this is in part a mere question of linguistic usage
and to that extent neither important nor interesting. The only substantial
question involved is whether explanation must be universal or may be con-
tingent. Nagel further shows, with examples (ten of them in his Chapter 2),
that contingent explanations are valid in any usual and proper sense of the
word "explanation." Nagel does not put the matter in just this way and he
makes other distinctions not pertinent in the present context, but in essence this
distinction between universal and contingent explanation parallels that be-
tween, on one hand, immanence and nonhistorical science, which involves
laws, and, on the other, configuration and historical science, which does not
involve laws but which does also have explanations.

The question "Why?" can be broken down into three others, each evoking
a different kind of explanation, as Pittendrigh (1958) and Mayr (1961), among
others, have discussed. "How?" is the typical question of the nonhistorical
sciences. It asks how things work: how streams erode valleys, how mountains
are formed, how animals digest food all in terms of the physical and chemical
processes involved. The first step toward explanation of this kind is usually a
generalized description, but answers that can be considered complete within
this category are ultimately expressed in the form of laws embodying invariable
relationships among variables. It is at this level that nonhistorical scientists
not only start but usually also stop.

The historical scientists nevertheless go on to a second kind of explanation
that is equally scientific and ask a second question, in the vernacular, "How
come?" How does it happen that the Colorado River formed the Grand
Canyon, that Cordilleras arose along the edge of a continent, or that lions live
on zebras? Again the usual approach is descriptive, the plain-story history of
changes in configurations, whether individual, as for the Grand Canyon, or
generalized to some degree, as for the concurrent evolution of lions and zebras.
This is already a form of explanation, but full explanation at this more complex
level is reached only by combination of the configurational changes with the
immanent properties and processes present within them and involved in those
changes. One does not adequately explain the Grand Canyon either by
describing the structure of that area and its changes during the Cenozoic or
by enumerating the physical and chemical laws involved in erosion, but by a
combination of the two.

There are two other kinds of scientific explanation to be mentioned here for
completeness, although they enter into geology only to a limited extent through
paleontology and are more directly biological and psychological. Both are
kinds of answers to the question "What for?" This question is inappropriate
in the physical sciences or the physical ("How?") aspects of other sciences,
historical or nonhistorical. "What does a stone fall for?" or "Why was the
Grand Canyon formed?" (in the sense of "What is it now for?") are questions

HISTORICAL SCIENCE 35

that make no sense to a modern scientist. Such questions were nevertheless
asked by primitive scientists (notably Aristotle) and are still asked by some
nonscientists and pseudoscientists. The rise of modern physical science re-
quired the rejection of this form of explanation, and physical scientists insisted
that such questions simply must not be asked. In their own sphere they were
right, but the questions are legitimate and necessary in the life sciences.

One kind of "What for?" question, for example, "What are birds 5 wings
for?" calls for a teleonomic answer. That they are an adaptation to flying is
a proper answer and partial explanation near the descriptive level. Fuller
explanation is historical: through a sequence of configurations of animals and
their environments wings became possible, had an advantageous function, and
so evolved through natural selection. Such a history is possible only in systems
with the elaborate feedback and information-storage mechanisms character-
istic of organisms, and this kind of explanation is inapplicable to wholly in-
organic systems (or other configurations). "What for?" may also be answered
ideologically in terms of purpose, explaining a sequence of events as means to
reach a goal. Despite Aristotle and the Neo-Thomists, this form of explanation
is scientifically legitimate only if the goal is foreseen. It therefore is applicable
only to the behavior of humans and, with increasing uncertainty, to some
other animals.

The question "How come?" is peculiar to historical science and necessary in
all its aspects. Answers to this question are the historical explanations. Never-
theless, the full explanation of history requires also the reductionist explanations
(nonhistorical in themselves) elicited by "How?" Teleonomic explanations
are also peculiar to historical science, but only to that part of it which deals
with the history of organisms.

Predictive Testing and Predictability

All of science rests on postulates that are not provable in the strictest sense.
The uniformity of the immanent, previously discussed, is only one such postu-
late, although perhaps the most important one for historical science. Indeed
it may be said that not only the postulates but also the conclusions of science,
including its laws and other theories, are not strictly provable. Proof in an
absolute sense occurs only in mathematics or logic when a conclusion is demon-
strated to be tautologically contained in axioms or premises. Since these
disciplines are not directly concerned with the truth or probability of axioms
or premises, and hence of conclusions drawn from them, their proofs are trivial
for the philosophy of the natural sciences. In these sciences, the essential point
is determination of the probability of the premises themselves, and mathe-
matics and logic only provide methods for correctly arriving at the implications
contained in those premises. Despite the vulgar conception of "proving a

36 GEORGE GAYLORD SIMPSON

theory," which does sometimes creep into the scientific literature, careful usage
never speaks of proof in this connection but only of establishment of degrees of
confidence.

In the nonhistorical sciences the testing of a proposition, that is, the attempt
to modify the degree of confidence in it, usually has one general form. A pos-
sible relationship between phenomena is formulated on the basis of prior
observations. With that formulation as a premise, implications as to phenomena
not yet observed are arrived at by logical deduction. In other words, a pre-
diction is made from an hypothesis. An experiment is then devised in order to
determine whether the predicted phenomena do in fact occur. The premise
as to relationships, the hypothesis, often has characteristics of a law, although
it may be expressed in other terms. As confidence increases (nothing contrary
to prediction is observed) it becomes a theory, which is taken as simultaneously
explaining past phenomena and predicting future ones.

Physical scientists (e.g. Conant, 1947) have often maintained or assumed
that this is the paradigm of testing ("verification" or increase of confidence)
for science in general. On this basis, some philosophers and logicians of science
(notably Hempel and Oppenheim, 1953) have concluded that scientific ex-
planation and prediction are inseparable. Explanation (in this sense) is a
correlation of past and present; prediction is a correlation of present and future.
The tense does not matter, and it is maintained that the logical characteristics
of the two are the same. They are merely two statements of the same relation-
ship. This conclusion is probably valid as applied to scientific laws, strictly
defined, in nonhistorical aspects of science. In previous terms, it has broad
perhaps not completely general validity for "How"? explanations. But we
have seen that there are other kinds of scientific explanations and that some of
them are more directly pertinent to historical science. It cannot be assumed
and indeed will be found untrue that parity of explanation and prediction is
valid in historical science.

Scriven (1959 and personal communication) has discussed this matter at
length. One of his points (put in different words) is that explanation and
prediction are not necessarily symmetrical, that in some instances a parity
principle is clearly inapplicable to them. Part of the argument may be para-
phrased as follows. If X is always preceded by A, A is a cause, hence at least
a partial explanation, of AT. But A may not always be followed by X. There-
fore, although A explains X when X does occur, it is not possible to predict the
occurrence of AT from that of A. A simple geological example (not from Scriven)
is that erosion causes valleys, but one cannot predict from the occurrence of
erosion that a valley will be formed. In fact, quite the contrary may occur;
erosion can also obliterate valleys.

The example also illustrates another point by Scriven (again in different
terms). The failure of prediction is due to the fact that erosion (A) is only a

HISTORICAL SCIENCE 37

partial cause of valleys (-Y). It is a (complex) immanent cause, and we have
omitted the configurational cause. Erosion is always followed by a valley
formation, A is followed by A', if it affects certain configurations. The total
cause, as in all historical events, comprises both immanent and configurational
elements. It further appears that prediction is possible in historical science,
but only to a limited extent and under certain conditions. If the immanent
causation is known and if the necessary similarities of configurational circum-
stances are known and are recurrent, prediction is possible.

The possibility of predicting the future from the past is nevertheless extremely
limited in practice and incomplete even in principle. There seem to be four
main reasons for these limitations. Mayr (1961) has discussed them in connec-
tion with historical aspects of biology, and with some modification his analysis
can be extended to historical science in general.

(1) A necessary but insufficient cause may not be positively correlated with
the usual outcome or event. This is related to the asymmetry of explanation
and prediction already discussed, and it is also discussed in other words by
Scriven (1959). Scriven's example is that paresis is caused by syphilis, but that
most syphilitics do not develop paresis. A modification of Mayr's example is
that mutation is a necessary cause for evolutionary change, but that such change
rarely takes the direction of the most frequent mutations. A geological example
might be that vulcanism is essential for the formation of basalt plateaus, but
that such plateaus are not the usual result of vulcanism.

(2) The philosophical interest of the foregoing reason for historical unpre-
dictability is reduced by the fact that the outcome might become predictable
in principle if all the necessary causes were known. But as soon as we bring in
configuration as one of the necessary causes, which must always be done in
historical science, the situation may become extremely, often quite impossibly,
complicated. Prediction is possible only to the extent that correlation can be
established with pertinent, abstracted and generalized, recurrent elements in
configurations. Considerations as to base level, slope, precipitation, and other
configurational features may be generalized so as to permit prediction that
a valley will be formed. It would be impossibly difficult to specify all the far
more complex factors of configuration required to predict the exact form of a
particular valley, an actual historical event. In such cases it may still be pos-
sible, as Scriven has pointed out in a different context, to recognize a posteriori
the configurational details responsible for particular characteristics of the actual
valley, even though these characteristics were not practically predictable. This
reason for unpredictability of course becomes more important the more complex
the system involved. As both Scriven and Mayr emphasize, it may become
practically insurmountable in the extremely complex organic systems involved
in evolution, and yet this does not make evolution inexplicable. Even in the
comparatively extremely simple physical example of the gas laws, it is obviously

38 GEORGE GAYLORD SIMPSON

impossible in practice and probably also in principle (because of the limitations
of simultaneous observation of position and motion) to determine the historical
configurations of all the individual molecules, so that the precise outcome of a
particular experiment is in fact unpredictable.

In this example the complications may be virtually eliminated and in his-
torical science they may often be at least alleviated by putting specification of
configurational causes on a statistical basis. This may, however, still further
increase the asymmetry of explanation and prediction. For instance, in Scriven's
previously cited example, as he points out, the only valid statistical prediction
is that syphilis will not produce paresis; in other words, that a necessary cause
of a particular result will not have that result. If, as a historical fact, a syphilitic
does become paretic, the event was not predictable even in principle. The point
is pertinent here because it demonstrates that a statistical approach does not
eliminate the effect of configurational complication in making historical events
unpredictable.

(3) As configurational systems become more complex they acquire charac-
teristics absent in the simpler components of these systems and not evidently
predictable from the latter. This is the often discussed phenomenon of emer-
gence. The classical physical example is that the properties of water may be
explicable but are not predictable by those of hydrogen and oxygen. Again the
unpredictability increases with configurational complications. It is difficult to
conceive prediction from component atoms to a mountain range, and to me,
at least, prediction from atoms to, say, the fall of Rome, is completely incon-
ceivable. It could be claimed that prediction of emergent phenomena would be
possible if we really knew all about the atoms. This might just possibly, and
only in principle, be true in nonhistorical science, as in the example of
2H+O >H 2 O. It would, however, be true in historical science only if we
knew all the immanent properties and also all the configurational histories of
all the atoms, which is certainly impossible in practice and probably in prin-
ciple. Whether or not the predictability of emergent phenomena is a philo-
sophical possibility (and I am inclined to think it is not), that possibility would
seem to have little heuristic and no pragmatic value.

(4) Scientific prediction depends on recurrence or repeatability. Prediction
of unique events is impossible either in practice or in principle. Historical
events are always unique in some degree, and they are therefore never pre-
cisely predictable. However, as previously noted, there are different degrees
of uniqueness. Historical events may therefore be considered predictable in
principle to the extent that their causes are similar. (This is a significant
limitation only for configurational causes, since by the postulate of uniformity
the immanent causes are not merely similar but identical.)

In practice, further severe limitations are imposed by the difficulties of
determining what similarities of cause are pertinent to the events and of ob-

HISTORICAL SCIENCE 39

serving these causal factors. It must also again be emphasized that such
prediction can only be general and not particular. In other words, prediction
does not include any unique aspect of the event, and in historical science it is
often the unique aspects that most require explanation. One might, for in-
stance, be able to provide a predictive explanation of mountain formation
(although in fact geologists have not yet achieved this) and also explanations
of the particular features of say, the Alps (achieved in small part), but the
latter explanations would not be predictive. (This is also an example of the
fact that unpredictive ad hoc explanation may be easier to achieve than pre-
dictive general explanation.)

At this point, one might wish to raise the question of what is interesting or
significant in a scientific investigation. In the physical study of gases or of
sand grains, the individuality (uniqueness) of single molecules or grains, slight
in any case, is generally beside the point. In dealing with historical events,
such as the formation of a particular sandstone or mountain range, individuality
often is just the point at issue. Here, more or less parenthetically, another aspect
and another use of the statistical approach are pertinent. A statistical descrip-
tion of variation in sand grains or of elevations, slopes, etc. in a mountain range
is a practical means of taking into account their individual contributions to the
over-all individuality of the sandstone or the mountain range.

Two other aspects of explanation and prediction in historical science may
be more briefly considered: the use of models, and prediction from trends and
cycles. Past geological events cannot be repeated at will, and furthermore,
prediction loses practical significance if, as is often the case in geology, its
fulfillment would require some thousands or millions of years. This is the
rationale for the experimental approach, using physical models to study the
historical aspects of geology and, when possible, other historical sciences. The
models abstract what are believed to be the essential general configurational
similarities of historical events (folding and faulting, valley erosion, and the
like) and scale these in space and time in such a way as to make them repeatable
at will and at rates that permit observation. With such models predictive
explanations can be made and tested. (The further problems of projecting from
model to geological space and time need not be considered here.)

Finally, the most common form taken by attempts at actual historical pre-
diction is the extrapolation of trends. In fact, this approach has no philosoph-
ical and little pragmatic validity. Its philosophical justification would require
that contingent causes be unchanging or change always in the same ways,
which observation shows to be certainly false. Its degree of pragmatic justi-
fication depends on the fact that trends and cycles do exist and (by definition)
continue over considerable periods of time. Therefore, at randomly distributed
times, established trends and cycles are more likely to continue than not.
Predictions through the extrapolation of trends are useful mostly for short

40 GEORGE GAYLORD SIMPSON

ranges of time; for larger ranges their likelihood decreases until the appropriate
statistical prediction becomes not continuation but termination or change of
trend or cycle within some specified time. The period of likely continuation or
justifiable extrapolation is, furthermore, greatly reduced by the fact that a
trend or cycle must already have gone on for a considerable time in order to be
recognized as actually existent. Present knowledge of geological and biological
(evolutionary) history suggests that all known trends and cycles have in fact
ended or changed except those which are now still within the span of likelihood
that is statistically indicated by the trends and cycles of the past. Moreover,
many supposed examples, such as regular cycles of mountain building or trends
for increase in size of machairodont sabers, now seem to have been mistaken.
Many real trends and cycles also turn out to be neither so uniform nor so long
continued as was formerly supposed, often under the influence of invalid his-
torical "laws" such as that of orthogenesis or of the pulse of the earth. It is
improbable that prediction about a total historical situation on this basis alone
is ever justified, even when prediction from causal properties and configurations
is possible within limits.

Strategy in Historical Science

The sequence hypothesis-prediction-experiment is not the only strategy of
explanation and testing in nonhistorical physical science. It is, however, so
often appropriate and useful there that philosophers who base their concepts
of science on physical science, as most of them do, tend to consider it ideal if
not obligatory. (On this point of view see, as a single example among many,
Braithwaite, 1953.) This is an example of the existing hegemony of the physical
sciences, which is not logically justifiable but has been fostered by human his-
torical and pragmatic factors. It has been shown that this strategy is also
possible in historical science, but that it here plays a smaller and less exclusive
role. It must be supplemented and frequently supplanted by other strategies.
These are in part implicit in what has already been said, but further notice of
some of the more important ones remains as the final aim of this essay. One
purpose is to demonstrate more fully and distinctly that nonpredictive explana-
tion and testing are in fact possible in historical geology and other historical
sciences.

The primary data of the historical scientist consist of partial descriptions of
configurations near the level of plain story. If the configurations are sequential
and connected, that is, if the later historically arose from the earlier, the ante-
cedent can be taken as including, at least in part, the configurational require-
ments and causes for the consequent. Even in such simple circumstances, a
direct causal connection can often be assumed on the basis of principles already
developed or on the basis of known parallels. For instance, partial configura-

HISTORICAL SCIENCE 41

tional causation is clearly involved in the sequence Hyiacotheriun (Eohippus)-
Orohippus or sand-sandstone. The latter example adds an important point: the
earlier configuration of a stratum now sandstone is not actually observed but
is inferred from the latter. The examples illustrate two kinds of explanatory
sequences available to the historical scientist. In one we have dated documents
contemporaneous with the events and so directly historical in nature and
sequence. In the other we have a pseudohistorical sequence such as that of
presently existing sands and sandstones. Their resemblances and differences
are such that we can be confident that they share some elements of historical
change, but that one has undergone more change than the other. In this case
it is easy to see that the sandstone belongs to a later period in the pseudohis-
torical sequence. One therefore infers for it a historically antecedent sand and
can proceed to determine what characteristics are inherited from that sand and
the nature of the subsequent changes.

The use of pseudohistorical sequences is another way in which the present
is a key to the past, but it does not involve another principle of uniformity.
The addition to the element of uniformity of immanent characteristics is
simply a descriptive resemblance or generalization of configuration applicable
to the particular case as a matter of observation. In practice, an historical
interpretation commonly involves both historical and pseudohistorical se-
quences. For example, study of the stratigraphy of a given region simultane-
ously concerns the directly historical sequence of strata and the history of each
stratum from deposition (or before) to its present condition as inferred on the
basis of appropriate pseudohistorical sequences.

A second form of strategy has a certain analogy with the use of multiple
experiments with controlled variables. The method is to compare different
sequences, either historical or pseudohistorical, that resemble each other in
some pertinent way. Resemblances in the antecedent configurations may be
taken to include causes of the consequent resemblances. It is not, however,
legitimate to assume that they are all necessary causes or that they include
sufficient causes. Even more important at times is the converse principle that
factors that differ among the antecedents are not causes of resemblances among
the consequents. By elimination when many sequences are compared, this may
warrant the conclusion that residual antecedent resemblances are necessary
causes. There is here applicable a principle of scientific testing in general:
absolute proof of a hypothesis or other form of inference is impossible, but
disproof is possible. Confidence increases with the number of opportunities for
disproof that have not in fact revealed discrepancies. In this application, con-
fidence that residual resemblances are causal increases with the number of
different sequences involved in the comparison. This form of strategy is ap-
plicable to most geological sequences, few of which are unique in all respects.
Obvious and important examples include the formation of geosynclines and

42 GEORGE GAYLORD SIMPSON

their subsequent folding, or many such recurrent phenomena as the strati-
graphic consequences of advancing and retreating seas.

An interesting special case arises when there is more resemblance among
consequent than among antecedent configurations: the phenomenon of con-
vergence. This has received much more attention in the study of organic
evolution than elsewhere, but nonorganic examples also occur. If I am cor-
rectly informed, the origin of various granites from quite different antecedents
is a striking example. Another fair example might be the formation of more
or less similar land forms by different processes: for example, the formation of
mountains by folding, faulting, or by erosion of a plateau; or the development
of plains and terraces by erosion or deposition. Doubtless most geologists can
find still other examples in their specialities. The special strategic interest of
convergence is that its elimination of noncausal factors often gives confidence
in identification of causes and increases knowledge of them. In organic evolu-
tion it has greatly increased understanding of the nature and limits of adapta-
tion by natural selection. In the example of the granites it shows that an essential
antecedent is not some one kind of lithology but atomic composition, and it
pinpoints the search for the processes bringing about this particular kind of
configuration of the atoms.

It has been previously pointed out that the explanation of an historical event
involves both configuration and immanence, even though the latter is not
historical in nature. Historical science therefore requires knowledge of the
pertinent immanent factors, and its strategy includes distinguishing the two
and studying their interactions. Nonhistorical science, by its primary concern
with the immanent, is the principal source of the historian's necessary knowledge
of immanent factors and his principal means of distinguishing these from con-
figurational relationships. A typical approach is to vary configurations in
experiments and to determine what relationships are constant throughout the
configurational variations. To a historical geologist, the function of a physical
geologist is to isolate and characterize the immanent properties of the earth
and its parts in that and other ways. The historical geologist is then interested
not in what holds true regardless of configuration, but in how configuration
modifies the action of the identified immanent properties and forces. In this
respect, the nonhistorical scientist is more interested in similarities and the
historical scientist in differences.

Here the historical scientist has two main strategies, both already mentioned.
They may be used separately or together. One proceeds by controlled experi-
mentation, in geology usually with scaled-down models although to a limited
extent experimentation with natural geological phenomena is also possible.
(The opportunities for experimentation are greater in some other historical
sciences.) The other might be viewed as complementary to the previously
discussed study of similarities in multiple sequences. In this strategy, attention

HISTORICAL SCIENCE 43

is focused on consequent differences, the causes of which are sought among the
observable or inferable differences of antecedent configurations. Although the
explanation is rarely so simple or so easy to identify, a sufficiently illustrative
example would be the presence in one valley and absence in another of a water-
fall caused by a fault, of a ledge of hard rock above shale, or of some other
readily observable local configuration.

Points always at issue in historical science are the consistency of proposed
immanent laws and properties with known historical events and the sufficiency
and necessity of such causation acting within known configurations. Probably
the strongest argument of the catastrophists was that known features of the
earth were inconsistent with their formation by known natural forces within
the earth's span of existence, which many of them took to be about 6000 years.
The fault of course was not with their logic but with one of their premises.
The same argument, with the same fallacy, was brought up against Darwin
when it was claimed that his theory was inadequate to account for the origin
of present organic diversity in the earth's span, then estimated by the most
eminent physicists as a few million years at most. Darwin stuck to his guns
and insisted, correctly, that the calculation of the age of the earth must be
wrong. Historical science has an essential role, both philosophical and prac-
tical, in providing such cross checks (mostly nonpredictive and nonexperi-
mental), both with its own theories and with those of other sciences as part of
the self-testing of science in general. A current geological example, perhaps
all the more instructive because it has not yet reached a conclusion, is the
controversy over continental drift and the adequacy of physical forces to bring
it about if indeed it did occur. (Incidentally the original motive for writing
Simpson, 1944, was to test the consistency and explanatory power of various
neontological theories of evolution by comparison with the historical record.)

The testing of hypothetical generalizations or proposed explanations against
a historical record has some of the aspects of predictive testing. Here, however,
one does not say, "If so and so holds good, such and such will occur," but,
"If so and so has held good, such and such must have occurred." (Again I
think that the difference in tense is logically significant and that a parity prin-
ciple is not applicable.) In my own field one of the most conspicuous examples
has been the theory of orthogenesis, which in the most common of its many
forms maintains that once an evolutionary trend begins it is inherently forced
to continue to the physically possible limit regardless of other circumstances.
This view plainly has consequences that should be reflected in the fossil record.
As a matter of observation, the theory is inconsistent with that record. A more
strictly geological example is the "pulse of the earth" theory, that worldwide
mountain-making has occurred at regularly cyclic intervals, which also turns
out to be inconsistent with the available historical data (Gilluly, 1949, among
others).

44 GEORGE GAYLORD SIMPSON

The study of human history is potentially included in historical science by
our definition. One of its differences from other branches of historical science
is that it deals with configurational sequences and causal complexes so exceed-
ingly intricate that their scientific analysis has not yet been conspicuously
successful. (Toynbee's (1946) correlation of similar sequences would seem to
be a promising application of a general historical strategy, but I understand
that the results have not been universally acclaimed by his colleagues.) A
second important difference is that so much of this brief history has been di-
rectly observed, although with varying degrees of accuracy and acuity and only
in its very latest parts by anyone whose approach can reasonably be called
scientific. Direct observation of historical events is also possible in geology and
other historical sciences, and it is another of their important strategies. Meticu-
lous observation of the history of a volcano (Paricutin) from birth to maturity
is an outstanding example. More modestly, anyone who watches a flash flood
in a southwestern arroyo or, for that matter, sees a stone roll down a hillside
is observing an historical event.

In geology, however, and in all historical science except that of human
history, the strategic value of observing actual events is more indirect than di-
rect. The processes observed are, as a rule, only those that act rapidly. The
time involved is infinitesimal in comparison with the time span of nonhuman
history, which is on the order of nlO 9 years for both" historical geology and
historical biology. Currently practicable resolution within that span varies
enormously but is commonly no better, and in some instances far worse,
than wlO 6 . The observed events are also both local and trivial in the great
majority of instances. They are in fact insignificant in themselves, but they are
extremely significant as samples or paradigms, being sequences seen in action
and with all their elements and surrounding circumstances observable. They
thus serve in a special and particularly valuable way both as historical (and
not pseudohistorical) data for the strategies of comparison of multiple sequences
and as natural experiments for the strategies of experimentation, including on
some but not all occasions that of prediction. (This, incidentally, is still a
third way in which the present is a key to the past, but again it involves
no additional uniformity principle.)

Direct observation of historical events is also involved in a different way in
still another of the historical strategies, that of testing explanatory theories
against a record. For example, such observations are one of the best means of
estimating rates of processes under natural conditions and so of judging whether
they could in fact have caused changes indicated by the record in the time
involved. Or the historical importance of observed short-range processes can
be tested against the long-range record for necessity, or sufficiency, or both.
An interesting paleontological example concerns the claims of some Neo-
Lamarckians who agree that although the inheritance of acquired characters

HISTORICAL SCIENCE 45

is too slow to be directly observed, it has been an (or the) effective long-range
process of evolution. The fossil record in itself cannot oifer clear disproof, but
it strips the argument of all conviction by showing that actually observed short-
range processes excluded by this hypothesis are both necessary and sufficient
to account for known history.

Conclusion

The most frequent operations in historical science are not based on the ob-
servation of causal sequences events but on the observation of results. From
those results an attempt is made to infer previous causes. This is true even when
a historical sequence, for example one of strata, is observed. Such a sequence
is directly historical only in the sense that the strata were deposited in a time
sequence that is directly available to us. The actual events, deposition of each
stratum, are not observable. In such situations, and in this sense, the present
is not merely a key to the past it is all we have in the way of data. Prediction
is the inference of results from causes. Historical science largely involves the
opposite: inferring causes (of course including causal configurations) from results.

The reverse of prediction has been called, perhaps sometimes facetiously,
postdiction. In momentary return to the parity of explanation and prediction,
it may be noted that if A is the necessary and sufficient cause of X and X is
the necessary and sole result of A, then the prediction of X from A and the
postdiction of A from AT are merely different statements of the same relationship.
They are logically identical. It has already been demonstrated and sufficiently
emphasized that the conditions for this identity frequently do not hold in
practice and sometimes not even in principle for historical science. Here, then,
postdiction takes on a broader and more distinct meaning and is not merely a
restatement of a predictive relationship. With considerable oversimplification
it might be said that historical science is mainly postdictive, and nonhistorical
science mainly predictive.

Postdiction also involves the self-testing essential to a true science, as has
also been exemplified although not, by far, fully expounded. Perhaps its
simplest and yet most conclusive test is the confrontation of theoretical explana-
tion with historical evidence. A crucial historical fact or event may be deduced
from a theory, and search may subsequently produce evidence for or against
its actual prior occurrence. This has been called "prediction," for example,
by Rensch (1960), sometimes with the implication that historical science is
true science because its philosophical basis does not really differ from that of
nonhistorical physics. The premise that the philosophy of science is necessarily
nonhistorical is of course wrong, but the argument is fallacious in any case.
What is actually predicted is not the antecedent occurrence but the subsequent
discovery; the antecedent is postdicted. Beyond this, perhaps quibbling, point,

46 GEORGE GAYLORD SIMPSON

the antecedent occurrence is not always a necessary consequence of any fact,
principle, hypothesis, theory, law, or postulate advanced before the postdiction
was made. The point is sufficiently illustrated on the pragmatic level by the
sometimes spectacular failure to predict discoveries even when there is a sound
basis for such prediction. An evolutionary example is the failure to predict
discovery of a "missing link," now known (Australopithecus), that was upright
and tool-making but had the physiognomy and cranial capacity of an ape.
Fortunately such examples do not invalidate the effectiveness of postdiction in
the sense of inferring the past from the present with accompanying testing by
historical methods. In fact the discovery of Australopithecus was an example of
such testing, for without any predictive element it confirmed (i.e. strengthened
confidence in) certain prior theories as to human origins and relationships and
permitted their refinement.

Another oversimplified and yet generally significant distinction is that his-
torical science is primarily concerned with configuration, and nonhistorical
science with immanence. Parallel, not identical, with this is a certain tendency
for the former to concentrate on the real and the individual, for the latter to
focus on the ideal and the generalized, or for both to operate with different
degrees of abstraction. We have seen, however, that interpretation and ex-
planation in historical science include immanence and, along with it, all the
facts, principles, laws, and so on, of nonhistorical science. To these, historical
science adds its own configurational and other aspects. When it is most charac-
teristically itself, it is compositionist rather than reductionist, examining the
involvement of primary materials and forces in systems of increasing complexity
and integration.

Historical science, thus characterized, cuts across the traditional lines between
the various sciences: physics, chemistry, astronomy, geology, biology, anthro-
pology, psychology, sociology-, and the rest. Each of these has both historical
and nonhistorical aspects, although the proportions of the two differ greatly.
Among the sciences named, the historical element plays the smallest role in
physics, where it is frequently ignored, and the greatest in sociology, where the
existence of nonhistorical aspects is sometimes denied one of the reasons that
sociology has not always been ranked as a science. It is not a coincidence that
there is a correlation with complexity and levels of integration, physics being
the simplest and sociology the most complex science in this partial list. Un-
fortunately philosophers of science have tended to concentrate on one end of
this spectrum, and that the simplest, so much as to give a distorted, and in some
instances quite false, idea of the philosophy of science as a whole.

Geology exhibits as even a balance of historical and nonhistorical elements
as any of the sciences, and here the relationships of the two may be particularly
clear. It is in a strategic position to illuminate scientific philosophy an
opportunity not yet sufficiently exploited.

HISTORICAL SCIENCE 47

Addendum

The preceding essay was read in manuscript and constructively criticized by a
number of geologists, mostly authors of other chapters in this work. The following
additional comments bear on points brought up by them.

A few critics objected to the term "immanent" on the grounds that it is unfamiliar
and is liable to confusion with "imminent," a word different in origin and meaning.
The most nearly acceptable substitute proposed was "inherent/ 5 which does not seem
to me equally precise or strictly appropriate in the intended sense. Since "immanent"
is here clearly defined and consistently used, I cannot believe that it will prove mis-
leading. It did not, in fact, mislead the readers who nevertheless criticized it.

Another critical suggestion was that under some special circumstances extrapolation
from historical trends may make unique events predictable. This is, I think, possible
to a limited extent and for relatively short-range prediction on a strictly probabilistic
basis, as is, indeed, pointed out in the preceding essay. The example given by the com-
mentator also illustrates the limitations: that the exhaustion of the preponderant part
of the fossil fuels within the next few centuries is predictable. In fact, contingent cir-
cumstances have changed so radically and unpredictably over recent years that the
term of this prediction has had to be greatly changed and has evidently become looser
and less reliable than was earlier believed. Even though some confidence may yet be
felt in the eventual outcome, such a historical prediction is on a different level from one
based on causal analysis apart from or in addition to trends.

Along similar lines, it was also remarked that prediction from cycles may be ex-
tremely reliable when the phenomena are definitely known to be cyclical, with planetary
motions as one example. That is, of course, true for the given example and for others
in which such current configurational changes as occur have come to be almost entirely
governed by cyclical immanent processes. The prediction is then based on the latter,
alone, and a truly historical element is limited. If a nonrecurrent historical change
should occur, for example if the mass of any planet were significantly altered, the pre-
dictions would prove false. To the extent that prediction is possible in such examples,
it depends on knowledge of immanent causes and on strictly recurrent configurations,
as specified above. Moreover, as this critic agrees in the main, similarly predictable
cycles may be discounted so far as geology is concerned.

Finally, radioactive decay of an isotope is cited as an example of a precisely pre-
dictable noncyclical phenomenon. I consider radioactive decay to be analogous to the
gas laws or to a chemical reaction; in each case the prediction of actual historical events
is not precise in principle, but the historical circumstances may be statistically so uniform
that changes in them can often be ignored in practice. Then the laws or the generalized
descriptions expressed in the appropriate equations hold good just to the extent that the
historical element can safely be ignored, and their predictions are not historical in
principle.

48 GEORGE GAYLORD SIMPSON

REFERENCES CITED

BARNHART, C. L., ed., 1948, The American college dictionary: New York and London,

Harper, 1432 pp.
BERNAL, J. D., 1951, The physical basis of life: London, Routledge and Kegan Paul,

80pp.
BRATTHWAITE, R. B. 3 1953, Scientific explanation: New York, Cambridge Univ. Press,

375 pp.

CON ANT, J. B., 1947, On understanding science: New Haven, Yale Univ. Press, 145 pp.
FARRAND, W. R., 1962, Frozen mammoths: Science, vol. 137, pp. 450-452. [See also

Lippman, 1962.]
GILUSPBE, C. C., 1951, Genesis and geology; a study in the relations of scientific thought,

natural theology, and social opinion in Great Britain, 1790-1850: Cambridge, Mass.,

Harvard Univ. Press, 315 pp.
GILLULY, J., 1949, Distribution of mountain building in geologic time: Gcol. Soc.

Am., B., vol. 60, pp. 561-590.
HEMPEL, C. G., and OPPENHEIM, P., 1953, The logic of explanation, in Readings in the

philosophy of science, H. Feigl and M. Brodbeck, eds.: New York, Appleton-Century-

Crofts, pp. 319-352.
HOBSON, E. W., 1923, The domain of natural science; The Gifford Lectures delivered in

the Univ. of Aberdeen in 1921 and 1922: New York, Macmillan, 510 pp.
HOOYKAAS, R., 1959, Natural law and divine miracle; a historical-critical study of the

principle of uniformity in geology, biology, and theology: Leiden, E. J. Brill, 237 pp.
LIPPMAN, H. E., 1962, Frozen mammoths: Science, vol. 137, pp. 449-450. [See also

Farrand, 1962.]
MAYR, E., 1961, Cause and effect in biology; kinds of causes, predictability, and

teleology are viewed by a practicing biologist: Science, vol. 134, pp. 1501-1506.
MORGAN, C. L., 1890-91, Animal life and intelligence: London, E. Arnold, 512 pp.
NAGEL, E., 1961, The structure of science; problems in the logic of scientific explana-
tion: New York, Harcourt, Brace, and World, 618 pp.
PTTTENDRIGH, C. S., 1958, Adaptation, natural selection, and behavior, in Behavior and

evolution, A. Roe and G. G. Simpson, eds.: New Haven, Yale Univ. Press, 557 pp.
RENSCH, B., I960, The laws of evolution, in Evolution after Darwin, S. Tax, ed., vol. 1:

Chicago, Univ. Chicago Press, pp. 95-116.
SCRIVEN, M., 1959, Explanation and prediction in evolutionary theory; satisfactory

explanation of the past is possible even when prediction of the future is impossible:

Science, vol. 130, pp. 477-482.
SIMPSON, G. G., 1944, Tempo and mode in evolution: New York, Columbia Univ.

Press, 237 pp.
, 1960, The history of life, in Evolution after Darwin, S. Tax, ed., vol. 1 : Chicago,

Univ. of Chicago Press, pp. 117-180.
, 1962, Notes on the nature of science by a biologist, in Notes on the nature of

science: New York, Harcourt, Brace, and World, pp. 7-12.

, 1963, Biology and the nature of science: Science, vol. 139, pp. 81-88.

TOYNBEE, A., 1946-47, A study of history: London, Oxford Univ. Press, 617 pp.

DAVID B. KITTS

University of Oklahoma

The Theory of Geology 1

Geologists have, throughout the history of their discipline, asked questions
about the nature of geology and its relationship to the other natural sciences.
Self-consciousness has been increasing among geologists during the past decade
as it has been increasing among all scientists. In print the most obvious sign
of an inclination toward self-examination is to be found in discussions of the
ideal geologic curriculum in colleges and universities. Many of the questions
raised in these discussions may be interpreted as questions about the theoretical
structure of the science.

Geologic Generalizations

In the discussion that follows, "geologic term" will be understood to mean
a term which fulfills a particular function in geology and is not a term necessary
for meaningful discourse outside geology. Geologic terms may have been in-
vented for the specific role which they fill, for example, "syncline"; or they
may be terms of the common language whose meaning has been expanded,
or more often restricted, to fill a particular technical role, for example, "sand."
A general statement which employs geologic terms in the above sense will be
called a "geologic generalization."

In a recently published paper (Kitts, 1963) I expressed the opinion that
geologic explanations are justified hi terms of generalizations which may be
compared to the laws of the other natural sciences. Most geologists have been
reluctant to attach the label "law" to any of the statements which they employ.
A few geologists, for example Bucher (1933), have used the term freely. It is

1 1 should like to express my gratitude to the members of the advanced seminar in
the history of science at the University of Oklahoma for their critical reading of the
manuscript. Carolyn Wares has given valuable assistance in the preparation of the
manuscript. Certain alterations in the last section of the paper have been made at
the suggestion of Dr. M. King Hubbert.

49

50 DAVID B. KITTS

not my intention to discuss the question of which statements should or should
not be called laws. This question has been discussed at some length by most
philosophers of science. For excellent recent discussions of the concept of a
scientific law, I refer the reader to Nagel (1961), Braithwaite (1953), and
Hempel and Oppenheim (1948). Although I shall have occasion to mention
current opinion on the kinds of statements which are to be called laws, my main
purpose here is to examine that great variety of general statements, whatever
we choose to call them, which serve as a part of the logical justification for vari-
ous kinds of geologic explanations.

Traditionally, scientific explanation has been regarded as a deductive opera-
tion. A generalization which functions in a deductive argument must be of
strictly universal form, that is, it must admit of no exceptions whatever. Con-
sequently, many writers have restricted the term "natural law" to statements
of strictly universal form. One can cite examples of geologic generalizations
which are universal. For example, the statement, "If the supply of new sand
is less than that moved over the crest ... the windward slope will be degraded
as the leeward slope grows, and thus the dune will migrate with the wind by
transfer of sand from one slope to the other" (Dunbar and Rodgers, 1957,
p. 20) is in form, though perhaps not in principle, universal.

Generalizations employing terms denoting probability or possibility, how-
ever, far outnumber generalizations of universal form in the geologic literature.
A striking feature of geologic discourse is the frequency with which such
words and phrases as "probably," "frequently," and "tends to," occur in
generalizations.

Explanations which contain these nonuniversal general statements are not
deductive, but rather inductive in form. As Hempel (1958, p. 40) has put it,
"This kind of argument, therefore, is inductive rather than strictly deductive
in character: it calls for the acceptance of E [a sentence stating whatever is
being explained] on the basis of other statements which constitute only partial,
if strongly supporting, grounds for it." Few geologists would disagree that the
generalizations which they employ seldom provide more than "partial, if
strongly supporting, grounds" for their conclusions.

It is customary to call general statements which admit of some exceptions
"statistical" or "probabilistic." According to Bunge (1961, p. 267) statistical
lawlike statements are "... propositions denoting regularities in the large or
in the long run. They contain logical constructs belonging to mathematical
statistics and characterizing central or overall trends ('average/ 'dispersion,'
'correlation,' and the like)." The generalization which states the rate of decay
of carbon fourteen is statistical in this sense.

General statements which contain terms denoting probability or possibility
but which are characterized by no precise formulation of statistical constructs
in numerical terms are better labeled "probabilistic" than "statistical." The
geologic literature abounds in statements of this sort. For example: "Wind-

THE THEORY OF GEOLOGY 51

driven sand, like snow, tends to settle in drifts in the wind shadow of topographic
obstructions." (Dunbar and Rodgers, 1957, p. 19)

Statistical inference is playing an increasingly important role in geologic
investigation and one cannot help being impressed by the frequency with
which statistical constructs appear in geologic literature. In much of this
literature, however, the precise statistical terminology- employed may obscure
the loosely probabilistic character of the generalizations containing it. In the
following quotation the antecedent portions of the generalization are framed
with statistical precision but the operators, for example "most" and "many,"
lack such precision.

Dune sands tend to be better sorted than river sands and a plot of standard
deviation (sorting) against mean grain-size indicates three fields, one for
river sands, one for dune sands, and a third field of overlap. This figure
points out that many river sands can be distinguished from dune sands and
vice versa on the basis of their textural parameters but that a wide field of
overlap exists. In practice this field of overlap is not necessarily a serious
matter, since most coastal, barrier bar, and lake dune sands have a standard
deviation of less than 0.40, and many desert and inland dune sands do not
exceed 0.50, whereas most river sands have a standard deviation in excess
of 0.50. (Friedman, 1961, p. 524, italics mine)

Scriven (1959) has recently suggested that the generalizations in terms of
which explanations of individual events are justified are usually neither uni-
versal nor statistical, nor even probabilistic, but rather belong to another
category of statements which he calls "normic statements." Scriven's concept
of the normic statement throws particular light on the problem of geologic
generalizations and explanations. He states (p. 464), "I suggest there is a
category of general statements, a hybrid with some universal features and some
statistical features, from which alone can be selected the role-justifying grounds
for good explanations of individual events," and (p. 466),

The statistical statement is less informative than the normic statement in a
very important way, although an exact statistical statement may be informa-
tive in a way a normic statement is not. The statistical statement does not
say anything about the things to which it refers except that some do and
some do not fall into a certain category. The normic statement says that
everything falls into a certain category except those to which certain special condi-
tions apply. And although the normic statement itself does not explicitly
list what count as exceptional conditions, it employs a vocabulary which
reminds us of our knowledge of this, our trained judgment of exceptions.

and (p. 466)

Now if the exceptions were few in number and readily described, one
could convert a normic statement into an exact generalization by listing

52 DAVID B. KITTS

them. Normic statements are useful where the system of exceptions, although
perfectly comprehensible in the sense that one can learn how to judge their
relevance, is exceedingly complex. We see immediately the analogy with
in fact, the normic character of physical laws. The physicist's training
makes him aware of the system of exceptions and inaccuracies, which, if
simpler, could be put explicitly in the statement of scope.

and finally (p. 467)

. . . statistical statements are too weak they abandon the hold on the
individual case. The normic statement tells one what had to happen in this
case, unless certain exceptional circumstances obtained; and the historical
judgment is made (and open to verification) that these circumstances did
not obtain.

Let us consider again the generalization from Dunbar and Rodgers (1957,
p. 19) concerning the accumulation of wind driven sand. "Wind-driven sand,
like snow, tends to settle in drifts in the wind-shadow of topographic obstruc-
tions." Offhand one would be inclined to regard this statement as probabilistic
because of the distinct probabilistic, or even statistical, connotation of the phrase
"tends to." But was it the intention of the authors to assert that if certain
specified conditions are realized then in a majority of cases certain other specified
conditions will follow, in which case the generalization must indeed be re-
garded as probabilistic; or was it their intention to assert that if certain specified
conditions are realized then certain other specified conditions will always
follow except where certain special conditions apply, in which case the general-
ization may be regarded as normic? I do not know what the intentions of the
authors were, but in the absence of this knowledge it seems to me fully as
plausible to regard the statement as normic as to regard it as probabilistic.
The question of meaning raised in connection with this generalization can be
raised in connection with a great number of seemingly probabilistic geologic
generalizations, and, for that matter, about a great number of seemingly
universal statements.

Most geologic generalizations, whatever their explicit form, could be re-
garded as normic statements, and the sense in which we actually understand
them is better conveyed by the term "normic" than by the terms "universal,"
"statistical," or "probabilistic." One thing is certain and that is the geologist
will not "abandon the hold on the individual case" by allowing statistical
statements to assume too important a role in his procedures, for, as I have
suggested earlier, it is the individual case which is his primary concern.

The normic character of some generalizations is made explicit. Whenever
the terms "normally" or "ordinarily" are encountered in a generalization it
can usually be assumed that the statement is normic. The following statement
is explicitly normic. "Carbonaceous material is a residue that remains after

THE THEORY OF GEOLOGY 53

various more or less mobile compounds, produced during the decomposition
of organic matter in an oxygen-deficient environment, have moved away.
Methane gas, for example, evolves and escapes under ordinary conditions."
(Weller, 1960, p. 152)

Scriven holds out the hope at least that normic statements can be converted
into universal statements. Von Engeln's (1942, p. 457) statement of the "law of
adjusted cross sections" represents an attempt to list "exceptional circum-
stances" which, if they should apply, invalidate the generalization.

Given that the surfaces of the joining glaciers are at the same level, that the
width of the main valley is not abruptly increased below the junction point,
and that the rate of motion of the main glacier is not greater below than
above the junction and all these conditions are met in numerous instances
it follows that there must be an abrupt increase in depth of the main valley
to accommodate the volume of the combined ice streams at the place where
they join.

It is clear that geologists tolerate a good deal more imprecision in their
generalizations than is technically necessary. Let us turn once again to the
generalization from Dunbar and Rodgers (1957, p. 19) on the accumulation
of wind-driven sand. It might be possible to specify the circumstances under
which wind-driven sand would always accumulate in the wind-shadow of
topographic obstructions. To accomplish this it would be necessary to specify
restricted ranges for the pertinent initial and boundary conditions such as the
velocity of the wind, the character and quantity of the sand, the form of the
obstruction, etc. The framing of universal statements is not, however, the
primary goal of the geologist. The primary goal is to frame general statements,
universal or not, on the basis of which explanations can be justified. The intro-
duction of specific initial and boundary conditions may permit the formulation
of strictly universal statements, but these statements will be of no use what-
soever unless these specific conditions can be independently determined. Very
often it is not possible to determine these conditions, and generalizations are
intentionally left in a loosely formulated state. This does not preclude the
possibility of taking certain specific conditions into account when they can be
independently inferred.

The view, not uncommon among geologists, that geology is as much an
"art" as a science, may stem from the fact that the system of exceptions as-
sociated with geologic generalizations is usually so complex that "judgment"
of their relevance may play an important role in investigation more im-
portant a role than in the natural sciences in which the system of exceptions
seems simpler.

The employment of many seemingly statistical and probabilistic generaliza-
tions imparts to geology an aspect of "indeterminacy." It would be very mis-

54 DAVID B. KITTS

leading to equate, as some geologists have done, this geologic indeterminacy
with the indeterminacy principle of modern physics. In modern quantum-
statistical mechanics indeterminacy is an integral part of theory. It is not
assumed that uncertainty can be removed by a more complete and detailed
specification of a particular set of variables. In geology, it seems to me, the
opposite assumption is usually made. Uncertainty is regarded as a feature to
be tolerated until more complete knowledge of variables allows its replacement
with certainty or, to put it another way, until probabilistic and normic state-
ments can be replaced by universal statements. Geologists are fundamentally
deterministic in their approach to scientific investigation. Indeterminacy, in the
sense that it is understood in physics, may enter into geology in those cases where
the principles of statistical mechanics enter directly into a geologic inference.

One final point should be made in connection with probability. Logicians
have called attention to the fact that there are two senses in which the term
"probability" may be used in connection with scientific hypotheses. In the
first sense it may be asserted that a particular hypothesis is probable or more
probable than some other hypothesis; for example, "It is probable that
glauconite marl represents the normal primary occurrence and that greensand
is a concentrate brought together like any other kind of sand during trans-
portation on the sea floor" (Dunbar and Rodgers, 1957, p. 184). In the second
sense a probability is assigned to an event within a hypothesis, for example,
"Almost all moving masses that begin as landslides in subaqueous situations
become mud flows." (Weller,- 1960, p. 158). Probabilistic terms are so fre-
quently and loosely employed by geologists that it is often difficult to determine
in which of the two senses a particular probabilistic term is to be understood.

Whether or not a geologic generalization is of universal form is certainly not
the only question which need concern us. A statement may be of strictly uni-
versal form but be restricted in scope. The scope of a statement corresponds
to the class of objects or events to which the statement applies. Every statement
is to some degree restricted in scope. Certain kinds of restrictions of scope can
seriously detract from the usefulness of a general statement which is intended
to have explanatory power. Among the most serious of these restrictions is
that imposed by reference to particular objects, or to finite classes of objects,
in particular spatio-temporal regions.

Generalizations which contain references to particular objects, or which
contain terms whose definition requires reference to particular objects are rare
in geology, as they are in other sciences. Generalizations in which reference is
made to a particular finite class of objects falling into a definite spatio-temporal
region are quite common and could conceivably cause some difficulty. In the
statement, "Major streams of the northern Appalachian region rise in the
Allegheny Plateau and flow southeastward to the Atlantic directly across the

THE THEORY OF GEOLOGY 55

northeast-southwest grain of rocks and structures ranging in age from Pre-
cambrian to Tertiary" (Mackin, 1938, p. 27), for example, it is clear that the
scope is restricted to a finite class of objects. Obviously the statement has no
explanatory power. We should not be justified in explaining the fact that a
particular major stream of the northern Appalachian region flowed southeast-
ward by reference to this statement, because, as Nagel (1961, p. 63) puts it, "if
a statement asserts in effect no more than what is asserted by the evidence for
it, we are being slightly absurd when we employ the statement for explaining or
predicting anything included in this evidence." No geologist would attempt to
explain anything at all by reference to this statement, nor did Mackin intend
that they should. The statement was obviously framed not to provide a means
of explanation and prediction, but rather to convey, as economically as possible,
information about some particular objects in a particular region. The statement
could have been formulated as a conjunction of singular statements without
change of meaning. Many general statements in geology whose scope is ex-
plicitly restricted are of this type.

As Nagel (1961, p. 63) has pointed out, a general statement which refers
to a group of objects or events which is presumably finite may be assigned an
explanatory role if the evidence for the statement is not assumed to exhaust
the scope of predication of the statement. The evidence cited to support the
statement, "As the velocity of a loaded stream decreases, both its competence
and its capacity are reduced and it becomes overloaded" (Dunbar and Rodgers,
1957, p. 9), for example, would consist of a finite number of observations, and
yet it is clearly not the intention of the authors to assert that this set of obser-
vations exhausts the scope of the statement.

A problem about the intention of the framer of a general statement might
arise. This problem is particularly likely to come up in connection with general
statements concerning the association of past events where the number of
instances cited in support of the statement is small. Do the instances cited to
support the following generalization exhaust its scope of predication or not?
"Fold mountains have their origin in the filling of a geosyncline chiefly with
shallow water sediments, conglomerates, sandstone, shales, and occasional
limestones." (von Engeln and Caster, 1952, p. 234) The class of fold mountains
contains a finite number of individuals. The question of scope cannot be an-
swered by an analysis of the statement, nor can it be answered by examining
the evidence for the statement. The answer lies in a determination of the
intention of the authors. Is it their intention merely to convey information
about a finite number of instances, or do they intend to assert that fold moun-
tains, past, present and future, have their origin in the filling of geosynclines?
If the latter is the case, then the intention is to assign an explanatory, and
possibly even a predictive, role to the statement. I cite this example to illustrate

56 DAVID B. KITTS

the importance of intention. I have no doubt that the authors meant to go
beyond a mere historical report.

It might be argued that all general geologic statements are restricted in
scope, because they contain implicitly or explicitly the individual name,
"the planet earth." Nagel (1961) regards Kepler's laws of planetary motion
as lawlike even though they contain individual names because, as he puts it
(p. 59), "The planets and their orbits are not required to be located in a fixed
volume of space or a given interval of time." I think that it is true of geologic
generalizations that the objects covered are not required to be associated with
"the planet earth." The generalization will hold wherever certain, to be sure
very specific, conditions obtain. In other words, it would be possible, in prin-
ciple at least, to express every term, including "the planet earth," in universal
terms and so eliminate essential reference to any particular object. The fact is,
of course, that no one worries about essential reference to the earth, and need
not worry so long as we practice our profession on earth.

A serious problem concerns suspected but unstated restriction of the temporal
scope of geologic statements. Because this problem stands at the very core of
any consideration of the uniformitarian principle, I shall discuss it in a separate
section of this paper.

The Theory of Geology

Is there any justification whatever for speaking about a "theory" of geology?
The answer to this question obviously hinges on another, namely, "What do
we mean by theory?" In an attempt to answer the latter question I shall quote
three philosophers of science on the subject.

Hempel (1958, p. 41) has this to say.

For a fuller discussion of this point, it will be helpful to refer to the familiar
distinction between two levels of scientific systematization: the level of
empirical generalization, and the level of theory formation. The early stages in
the development of a scientific discipline usually belong to the former level,
which is characterized by the search for laws (of universal or statistical form)
which establish connections among the directly observable aspects of the
subject matter under study. The more advanced stages belong to the second
level, where research is aimed at comprehensive laws, in terms of hypo-
thetical entities, which will account for the uniformities established on the
first level.

Nagel (1961, p. 83) recognizes the usefulness of a distinction between
"theoretical laws" and what he calls "experimental laws," but calls attention
to the difficulties which may arise in the attempt to label a particular law as

THE THEORY OF GEOLOGY 57

one or the other. He points out, however, that,

Perhaps the most striking single feature setting off experimental laws from
theories is that each "descriptive" (i.e. nonlogical) constant term in the
former, but in general not each such term in the latter, is associated with at
least one overt procedure for predicating the term of some observationally
identifiable trait when certain specified circumstances are realized. The
procedure associated with a term in an experimental law thus fixes a definite,
even if only partial, meaning for the term. In consequence, an experimental
law, unlike a theoretical statement, invariably possesses a determinate
empirical content which in principle can always be controlled by observa-
tional evidence obtained by those procedures.

And finally Carnap (1956, p. 38), emphasizing the linguistic consequences
of the distinction made by Nagel and Hempel, states:

In discussions on the methodology of science, it is customary and useful
to divide the language of science into two parts, the observation language
and the theoretical language. The observation language uses terms desig-
nating observable properties and relations for the description of observable
things or events. The theoretical language, on the other hand, contains
terms which may refer to unobservable events, unobservable aspects or
features of events, e.g., to microparticles like electrons or atoms, to the
electromagnetic field or the gravitational field in physics, to drives and
potentials of various kinds in psychology, etc.

We may now proceed to examine geology with these "theoretical" charac-
teristics in mind. The claim has been repeatedly made that geology was in the
past, and to some extent remains today, "descriptive." For example, Leet and
Judson (1954, p. ii) state, "Originally geology was essentially descriptive, a
branch of natural history. But by the middle of the 20th century*, it had de-
veloped into a full-fledged physical science making liberal use of chemistry,
physics and mathematics and in turn contributing to their growth." Just what
is meant by "descriptive" in this sort of statement? It certainly cannot be taken
to mean that geology is today, or has been at any time during the last two
hundred years, wholly, or even largely, concerned with the mere reporting of
observations. The very fact that during this period geologists have remained
historical in their point of view belies any such contention. The formulation
of historical statements requires inferential procedures that clearly go beyond
a mere "description."

The feature that has apparently been recognized by those who characterize
geology as descriptive, and indeed the feature which we can all recognize, is
that most geologic terms are either framed in the observation language Or can
be completely eliminated from any geologically significant statement in favor

58 DAVID B. KITTS

of terms that are so framed. This is simply to say that most geologic terms are
not theoretical. Even Steno's four great "axioms" may be regarded as "ob-
servational" rather than "theoretical."

The abundance of historical terms in geology may give rise to some confusion
when an attempt is made to distinguish between observation terms and theo-
retical terms. Historical terms involve some kind of historical inference to a
past event or condition and consequently require for their definition reference
to things which have not been observed by us. Consider, for example, the term
"normal fault" which has been defined as a fault "in which the hanging wall
has apparently gone down relative to the footwall" (Billings, 1954, p. 143).
The label "theoretical" would probably be denied this term by most philos-
ophers of science because the past event can be described in the observation
language and could consequently, in principle at least, be observed.

The question then arises, are there any geologic terms which clearly qualify
as theoretical? This is a difficult question because, as many authors have
pointed out, the distinction between theoretical terms and observation terms is
far from clear. There are certainly no geologic terms so abstract as the higher-
level theoretical constructs of quantum mechanics, for example. On the other
hand there are a number of geologic terms which almost certainly qualify as,
what might be called, "lower-level theoretical terms," that is terms of rela-
tively limited extension. "Geosyncline," it seems to me, is such a term. Ac-
cording to Kay (1951, p. 4) a geosyncline may be defined as "a surface of
regional extent subsiding through a long time while contained sedimentary and
volcanic rocks are accumulating; great thickness of these rocks is almost in-
variably the evidence of the subsidence, but not a necessary requisite. Geo-
synclines are prevalently linear, but nonlinear depressions can have properties
that are essentially geosynclinal."

Another candidate for the title "theoretical term" is "graded stream."
According to Mackin (1948, p. 471),

A graded stream is one in which, over a period of years, slope is delicately
adjusted to provide, with available discharge and with prevailing channel
characteristics, just the velocity required for the transportation of the load
supplied from the drainage basin. The graded stream is a system in equi-
librium; its diagnostic characteristic is that any change in any of the con-
trolling factors will cause a displacement of the equilibrium in a direction
that will absorb the effect of the change.

If I am correct in regarding these terms as theoretical, we might expect that
some difficulties would arise if an attempt were made to define these expressions
in terms of observables. I do not, however, feel inclined or qualified to go into
the problem of "rules of correspondence" at this point.

THE THEORY OF GEOLOGY 59

What is the function of theoretical terms in a scientific system? The answer
to this question was given, I think, in the passage from Hempel (1958, p. 41)
quoted earlier, and in the words of Nagel (1961, pp. 88-89) when he says,

An experimental law is, without exception, formulated in a single state-
ment; a theory is, almost without exception, a system of several related
statements. But this obvious difference is only an indication of something
more impressive and significant: the greater generality of theories and their
relatively more inclusive explanatory power.

The general statements which contain the terms "geosyncline" and "graded
stream" have, relative to other geologic statements, great generality and quite
clearly fulfill the function which theoretical terms are designed to fill, that is,
they establish some considerable "explanatory and predictive order among
the bewildering complex 'data' of our experience, the phenomena that can be
directly observed by us." (Hempel, 1958, p. 41)

One of the most striking features of scientific systems with a highly developed
theoretical structure is the degree to which it is possible to make logical con-
nections between the various general statements contained in the system. A
given general statement may, in combination with other general statements or
even with singular statements, serve as the basis for the deductive derivation of
other general statements, or itself be derivable from other generalizations in the
same way. It is this feature which provides empirical systems with their sys-
temicity. The importance of system is indicated by Braithwaite (1953, pp. 301-
302), who holds that a hypothesis is to be considered "lawlike" only on the
condition that it "either occurs in an established scientific deductive system
as a higher-level hypothesis containing theoretical concepts or that it occurs in
an established scientific deductive system as a deduction from higher-level
hypotheses which are supported by empirical evidence which is not direct
evidence for itself."

As Braithwaite suggests, in the case of a generalization (G) which is deduc-
tively related to other generalizations, it is possible to distinguish between
indirect and direct supporting evidence for it. Any empirical evidence which
supports a generalization which is deductively related to G will count as indirect
evidence in support of G. Nagel (1961, p. 66) emphasizes the importance of
indirect evidence in the statement, "Indeed, there is often a strong disinclina-
tion to call a universal conditional L a 'law of nature,' despite the fact that it
satisfies the various conditions already discussed, if the only available evidence
for L is direct evidence."

How much systematization is there in the body of geologic knowledge? If
we consider geologic generalizations, that is, general statements containing
only special geologic terms, we are immediately struck with how little system-
atization obtains. There is some, of course, and it is usually provided in terms

60 DAVID B. KITTS

of such unifying concepts as "graded stream" and "geosyncline." This low-
degree systematization is not surprising, for, as every elementary geology text
insists, geology is not a discipline unto itself. We have only to introduce into
our system the generalizations and laws of physical-chemical theory and the
logical connections within the branches and between the branches of geology
become more impressive. It may still be a matter of contention among geolo-
gists as to whether every geologic generalization has been, or in principle could
be, incorporated into a theoretical system in terms of physical-chemical laws.
If a geologic generalization is truly independent, however, no matter how useful
it may be, its status will suffer because no indirect evidence can be presented
to support it, and we are likely to speak of it as "merely" an empirical general-
ization.

The question is not whether geology is chemistry and physics, but whether
or not geologists will utilize physical-chemical theory with all its admitted
imperfections and all its immense power. - The question has been answered.
Almost all geologists proceed as if every geologic statement either has now, or
eventually will have, its roots in physical-chemical theory. Geologic theory is
modified to accommodate physical-chemical theory and never conversely.
Furthermore, this confidence in physics and chemistry is not a unique feature of
twentieth-century geology, for Hutton (1795, pp. 31-32) said:

It must be evident, that nothing but the most general acquaintance with
the laws of acting substances, and with those of bodies changing by the
powers of nature, can enable us to set about this undertaking with any
reasonable prospect of success; and here the science of Chemistry must be
brought particularly to our aid; for this science, having for its object the
changes produced upon the sensible qualities, as they are called, of bodies,
by its means we may be enabled to judge of that which is possible according
to the laws of nature, and of that which, in like manner, we must consider as
impossible.

Certainly no consistent, economical, complete deductive system of geology
exists, but I think that we can detect a suggestion of such a system. In this
system the higher-level universals, or postulates, which serve as the basis for
the deductive derivation of other generalizations and of singular statements of
the system are exactly those universals which serve this purpose in physics and
chemistry. The theory of geology is, according to this view, the theory of
physics and chemistry. The geologist, however, unlike the chemist and the
physicist, regards this theory as an instrument of historical inference.

The generalizations containing the theoretical terms "geosyncline" and
"graded stream" are not regarded as "fundamental laws of nature" nor as
postulates of a theory of geology. Geologists are committed to the view that

THE THEORY OF GEOLOGY 61

generalizations containing these terms can be derived from higher-level gen-
eralizations which belong to physics. Thus Leopold and Langbein (1962,
p. A 11) state with reference to the graded-stream state, "A contribution made
by the entropy concept is that the Equilibrium profile 5 of the graded river is
the profile of maximum probability and the one in which entropy is equally
distributed, constituting a kind of isentropic curve."

The activity of attempting to relate geologic generalizations to physical-
chemical theory is not a game played by geologists for its own sake, nor is it
played solely for the sake of increasing our confidence in these generalizations
by providing indirect supporting evidence for them. It is the hope of those
engaged in this activity, which I am tempted to call "theoretical geology,"
that putting a geologic generalization into a theoretical setting will allow a
more precise and rigorous formulation of it, and thus may increase its utility.
This consideration is undoubtedly behind Leopold and Langbein's statement
(1962, p. A 1), "The present paper is an attempt to apply another law of
physics to the subject for the purpose of obtaining some additional insight into
energy distributions and their relation to changes of land forms in space and
time."

A major objection to the view that geology is "descriptive" and thus in some
sense not "mature" is that the scope of geologic knowledge and theory is in
no sense now, nor has it ever been, exhausted by what can be framed in special
geologic terms and in terms of the common language alone. The laws which
provide for much of the higher-level systematization in geology are perhaps
not immediately associated with a theory of geology because they are laws
without particular geologic reference which are ordinarily recognized as
belonging to the theory of chemistry and physics. To consider the science of
geology as composed of just those statements containing particular geologic
references amounts to a wholly artificial decapitation of a rather highly organ-
ized theoretical structure. An arbitrary distinction among physical, chemical,
and geologic terms is not significant here. What is significant is the role played
by laws and terms, no matter what their origin or what we may choose to call
them, in the theory of geolog> r . Modern geology assumes all of contemporary
physical-chemical theory and presents on the basis of this assumption a high
degree of logical integration.

It has been suggested, for example by Leet and Judson (1954, p. ii) in the
statement quoted above, that geology not only draws upon physical-chemical
theory but may contribute to this theory. In some sense, at least, this is true
since geologic problems may serve to stimulate investigations which are pri-
marily physical or chemical in nature, particularly if the problems arise in such
fields as geochemistry, geophysics, and crystallography in which the borderline
between geology and the rest of physical science is difficult to draw. Tradi-
tionally, however, geologists have not tampered with nongeologic theory in

62 DAVID B. KITTS

any very direct way. The physical-chemical theory of their time is a standard
which tends to be accepted by each generation of geologists.

The Origin of Geologic Generalizations

It has often been said that geology is "inductive." It may be that, in some
cases at least, the application of this label is meant to convey the fact that the
explanations proposed by geologists are probabilistic rather than deductive.
Usually, however, I think that the term "inductive" when applied to geology-
is meant to convey that the generalizations of geology are to be regarded as
inductive generalizations. If we agree that most geologic generalizations are
not theoretical, then this latter view of the inductive nature of geology is en-
tirely plausible since, as Nagel (1961, p. 85) has pointed out, "An immediate
corollary to the difference between experimental laws and theories just dis-
cussed is that while the former could, in principle, be proposed and asserted
as inductive generalizations based on relations found to hold in observed data,
this can never be the case for the latter." The problem of the process by which
theoretical laws are formulated is far beyond the scope of this paper.

Although the usual procedure in geology is to explain a generalization which
has been formulated inductively in terms of higher-level generalizations, using
the laws of chemistry and physics, it is possible to start with these higher-level
generalizations and deduce from them generalizations of geologic significance
and utility. We thus speak of the deduction of the mineralogical phase rule
from the phase rule of Gibbs (see, for example, Turner, 1948, p. 45). It is my
belief that the most dramatic advances in geologic theory are to be expected
from this deductive method.

The Uniformi tarian Principle

There is widespread agreement among geologists that some special principle
of uniformity is a fundamental ingredient of all geologic inference. Longwell
and Flint (1955, p. 385) go so far as to say that, "The whole mental process
involved in this reconstruction of an ancient history is based on that cornerstone
of geologic philosophy, the principle of uniformitarianism, probably the great-
est single contribution geologists have made to scientific thought." Despite
this general agreement about the importance of the principle, geologists hold
widely divergent views as to its meaning. So divergent are these views, in fact,
that one is led to conclude that there has been little or no resolution of the
problems which gave rise to the famous controversies between the "uniformi-
tarians" and the "catastrophists" in the nineteenth century. Though the prob-
lems have not been solved, the controversy has subsided. A number of accounts
of the history of the uniformitarian principle are available, including an excel-
lent recent one by Hooykaas (1959).

THE THEORY OF GEOLOGY 63

There are two principal problems concerning the concept of uniformity.
The first problem concerns the grounds upon which the assertion of uniformity
rests. The second problem concerns the precise nature of the restriction which
it imposes.

As to the grounds upon which the assertion of uniformity rests, it would be
difficult, if not impossible, to contend that any meaningful empirical support
can be presented for the view that "geologic process" in the past was in any
way like this process in the present, simply because statements about process
must be tested in terms of statements about particulars and no observation of
past particulars can be made. We do, of course, formulate statements about
particular past conditions. These statements are not supported by direct
observations, however, but rather are supported by inferences in which some
assumption of uniformity is implicit. In terms of the way a geologist operates, there
is no past until after the assumption of uniformity has been made. To make statements
about the past without some initial assumption of uniformity amounts, in
effect, to allowing any statement at all to be made about the past. Geologists,
as I have suggested earlier, are primarily concerned with deriving and testing
singular statements about the past. The uniformitarian principle represents,
among other things, a restriction placed upon the statements that may be
admitted to a geologic argument, initially at the level of primary historical
inference. Once the assumption of uniformity has been made for the purposes
of primary historical inference, then it may be possible to "demonstrate" some
uniformity or, for that matter, depending upon how strictly the principle is
applied in the first place, some lack of uniformity. The essential point is that
the assumption of uniformity must precede the demonstration of uniformity
and not vice versa. In principle, at least, it would be possible to make at the
outset some assumptions about particular conditions at some time in the past
and on the basis of these assumptions "test" whether some law held at that
time. Geologists have chosen to do the opposite, however, by making some
assumptions about the uniformity of the relationships expressed in a law and
on this basis to derive and test some singular statements about the past.

The view that the uniformity of "nature" or of "geologic process" represents
an assumption that is made in order to allow geologists to proceed with the
business of historical inference is a very old one. In Hutton (1788, p. 301-302),
for example, we find this view expressed in the statement, "We have been
representing the system of this earth as proceeding with a certain regularity,
which is not perhaps in nature, but which is necessary for our clear conception
of the system of nature." And Lyell (in a letter to Whewell, 1837, from "Life,
Letters and Journals," 1881, Vol. II, p. 3) expresses a similar view when he
states,

The former intensity of the same or other terrestrial forces may be true; I
never denied its possibility; but it is conjectural. I complained that in at-
tempting to explain geological phenomena, the bias has always been on the

64 DAVID B. KITTS

wrong side; there has always been a disposition to reason a priori on the
extraordinary violence and suddenness of changes, both in the inorganic
crust of the earth, and in organic types, instead of attempting strenuously
to frame theories in accordance with the ordinary operations of nature.

The uniformitarian principle is not usually formulated as an assumption by con-
temporary geologists.

If the uniformitarian principle is to be regarded as a methodological device
rather than an empirical generalization, a reasonable step toward a solution
of the problems surrounding the principle would be an attempt to explicate it.
"What statements in the theory of geology shall be regarded as untimebound"?
is a question which might be asked in connection with such an attempt.

It is plausible to regard every singular descriptive statement in geology as
timebound, to some degree at legist, simply because a principal aim of geologic
endeavor is to "bind" singular descriptive statements temporally, which is
another way of saying that a principal aim of geologic endeavor is to produce
historical chronicle. Singular descriptive statements may apply to different
times and different places, but it cannot be assumed that they apply to all
times and to all places, and consequently their uniformity cannot be assumed
for the purposes of inference.

There are few geologists who would disagree with this view regarding
singular descriptive statements, and for this reason I may be accused of pro-
pounding the obvious. Before the accusation is made, however, let us consider
a passage from Read (1957, p. 26).

This difficulty confronting static metamorphism has been tackled by Daly,
who meets it by relaxing the rigidity of the doctrine of uniformitarianism.
He admits that, compared with its proposed potency in the Pre-Cambrian
times, load metamorphism must have been of relatively little importance in
later geological eras. To account for this, he assumes that the earth's thermal
gradient was steeper during the formation of the Pre-Cambrian so that
regional metamorphism under a moderate cover was possible. He considers
that this speculation concerning a hotter surface to the earth is c no more
dangerous than the fashionable explanation of all, or nearly all, regional
metamorphism by erogenic movements.' I agree that. . . [although] uni-
formitarianism suits the events of the 500 million years of geological history
as recorded in the Cambrian and later fossiliferous rocks, it may quite likely
not be so valid for the 2,000 million years of Pre-Cambrian time.

To suggest that the thermal gradient of the earth was steeper during the Pre-
cambrian than it is today is to make an assertion about particular conditions
during the past. This assertion is perhaps not so soundly based as many others,
but it has resulted from the kind of retrodictive inference that geologists fre-
quently perform. To claim that this statement requires a relaxation of the

THE THEORY OF GEOLOGY 65

doctrine of uniformitarianism is, in principle, equivalent to claiming that an
assertion that the topography of Oklahoma during Permian time was different
from what it is today requires a similar relaxation.

The statements that the geologist wishes to regard as untimebound are of
general form. This is revealed by the fact that he speaks of uniformity of
"cause" or "process" or "principle" or "law," concepts which are expressed
in general statements. The whole problem of the strictness with which the
uniformitarian assumption is to be applied revolves ultimately around the
question of which of the many statements of general form in the theory of
geology are to be regarded as being without temporal restriction.

Geologists usually speak of the problem of uniformitarianism as though it
were a problem of whether or not the relationships expressed in scientific laws
could be regarded as constant throughout geologic time. Is this really the
problem? Are geologists bothered, for example, by the question of whether or
not the relationship expressed in the equation F = G m^m^/d* can be regarded
as independent of time? They may occasionally speculate about the "uniform-
ity of nature," but such speculations do not enter into the conduct of geologic
investigation in any significant way. Here again it is to the authority of physical-
chemical theory that appeal is made. If an expression is untimebound in this
theory, it is untimebound in the theory of geology, particularly if the expression
is regarded as a "fundamental law." Furthermore, if a law should be formu-
lated in which the expressed relationship were a function of the passage of
historic time, and if this law should come to be regarded as a valid part of
some physical or chemical theory, there would probably be no reluctance on
the part of geologists to accept the law nor would the acceptance signal an
abandonment of the uniformitarian principle. The view that it is "law"
which is to be regarded as uniform has been suggested many times before, for
example in the following statement from Moore (1958, p. 2): "As foundation,
we accept the conclusion that nature's laws are unchanging."

Not only is the most general form of a fundamental law regarded as untime-
bound but so are substitution instances and other logical consequences of such
a law. We should agree, for example, that at any time and place where two
spherical masses of one pound each and of uniform density are held with their
centers one foot apart, the force of attraction between them will be 3.18 X
10"" 11 pounds weight. We do not conclude from this, however, that the attrac-
tion between any two objects at any time is 3.18 X 10~~ n pounds weight, for to
do so would necessitate the assumption of the temporal uniformity of certain
particular conditions. This particular attractive force is to be inferred only
where the specified antecedent conditions actually obtain. That these particular
conditions, or for that matter any particular conditions, did in fact obtain at
a particular time and place is a matter for independent verification. The law
does not "change," but different substitution instances of the law are applicable

66 DAVID B. KITTS

to different times and places. The immediate problem which the geologist
faces is not one of uniformity, but one of determining which, if any, of the
infinite number of substitution instances of a general law is applicable in a
particular case.

Let us consider a geologic statement in which doubt about temporal exten-
sion is expressed. "Probably only time and the progress of future studies can
tell whether we cling too tenaciously to the uniformitarian principle in our
unwillingness to accept fully the rapid glacial fluctuations as evidenced by
radiocarbon dating." (Horberg, 1955, p. 285) The author has suggested, has
he not, that there was available a generalization concerning the rate of glacial
fluctuation, perhaps supported by observations of extant glaciers, whose
validity had become doubtful because of our greater confidence in another
generalization. Isn't there something suspicious about a statement which is
supposed to have some degree of temporal extension and which contains a
reference to a particular rate or even to a limited range of rates? There are
many physical laws in the form of equations which allow the calculation of a
particular rate when particular values are substituted for variables. The
unsubstituted form of a physical equation, however, contains no reference to
a particular rate. Rate of glacial fluctuation does not remain constant through-
out time any more than the force of attraction between objects remains con-
stant throughout time. Rate of glacial advance and retreat is dependent upon
a number of variables, among which are the topography, the thickness of the
ice, and the temperature of the atmosphere. Unless specific values can be sub-
stituted for each of these variables a law of glacial movement cannot be mean-
ingfully applied at all, and could certainly not be called upon to serve as the
basis for the assertion that the rate of glacial movement had some uniformity
in geologic time. Horberg's problem did not involve a question of the "uni-
formity of nature" but rather it involved a question of whether or not he was
in a position to determine particular values for each of the variables upon which
rate of glacial movement might be presumed to be dependent. No explication
of the concept of "the uniformity of nature" nor "universal causation" could
have served as a basis for the solution of this problem.

Another problem presents itself at this point. Is there a law of glacial move-
ment which has been formulated with sufficient precision to allow us to say
what the pertinent variables are, let alone determine specific values for them?
The answer would have to be that there is not. What is available is an im-
precisely formulated generalization, probably normic in form, which may
serve as the basis for a variety of inferences but cannot serve as a basis for a
calculation of the rate of glacial advance and retreat during a particular interval
of time. The temporal universality of imprecisely formulated probabilistic and
normic generalizations will always be suspect simply because it is character-
istic of such statements that they contain what must be regarded as hidden
references to particulars. We are able to increase our confidence in the temporal

THE THEORY OF GEOLOGY 67

universality of a generalization by formulating it with greater precision. It
must be borne in mind, however, that progress in geology- has depended upon
the willingness of geologists to assume the temporal uniformity of a great
variety of imprecisely formulated generalizations.

There is a long-standing philosophical problem concerning the uniformity
of nature. It has been discussed at great length in the past and it will be dis-
cussed in the future. This problem is everybody's problem, not a special prob-
lem for geologists. The special problem for geologists concerns first the avail-
ability of appropriate laws and second the applicability of laws to particular
situations. In an attempt to solve the problem, many geologists are engaged
in work that is directed toward the goal of increasing the precision with which
geologic generalizations are formulated. They proceed by observing in the
field, experimenting in the laboratory and, increasingly, by paper and pencil
operations within a theoretical framework. The latter procedure may turn
out to be the most satisfying of all because never are we so confident in the
precision of a law as when it has been made a part of some theory.

The precise formulation of geologic generalizations cannot be expected to
lead to an immediate solution of all the problems facing the contemporary
geologist. The principal use to which geologic generalizations are put is as a
basis for explanatory inferences which allow the derivation and testing of
singular statements about the past. A generalization cannot be used in this
way unless specific values can be substituted for most of the variables contained
in it. A specific value which is substituted for a variable in a generalization
employed in a retrodictive inference must have been determined on the basis
of another retrodictive inference which is itself dependent upon some general-
ization. The necessary interdependence of the bewildering variety of geologic
retrodictive inferences upon one another seems, at times, to result in a piling
of uncertainty upon uncertainty. Remarkably, however, confidence in geologic
statements about the past as reliable descriptions of particular conditions, in
particular places, at particular times, is high. The basis of this confidence is
that not only are retrodictive inferences to a large extent dependent upon one
another, but that they serve as the only means of verification of one another.
Each statement about the past is tested against other statements about the past.
Laboriously we build the chronicle, selecting, eliminating, and modifying as
we proceed, bringing to bear, in addition to an immensely complicated infer-
ential apparatus, our trained judgment.

REFERENCES CITED

BILLINGS, M. P., 1954, Structural geology: New York, Prentice-Hall, 514 pp.
BRArrHWAiTE, R. B., 1953, Scientific explanation: New York, Cambridge Univ. Press,

375 pp.
BUCHER, W. H., 1933, The deformation of the earth's crust; an inductive approach to

the problems of diastrophism: Princeton, Princeton Univ. Press, 518 pp.

68 DAVID B. KITTS

BUNGE, M., 1961, Kinds and criteria of scientific laws: Phil. Sci. 3 vol. 28, pp. 260-281.

CARNAP, R. 3 1956, The methodological character of theoretical concepts, p. 38-76 in
H. Feigl and M. Scriven, eds., Minnesota Studies in the Philosophy of Science,
vol. 1. The foundations of science and the concepts of psychology and psychoanalysis:
Minneapolis, Univ. Minnesota Press, 346 pp.

DUNBAR, C. O., and RODGERS, J., 1957, Principles of stratigraphy: New York, John
Wiley, 356 pp.

FRIEDMAN, G. M., 1961, Distinction between dune, beach, and river sands from their
textural characteristics: J. Sediment. Petrology, vol. 31, pp. 514-529.

HEMPEL, G. G., 1958, The theoretician's dilemma; a study in the logic of theory con-
struction, pp. 37-98 in H. Feigl, M. Scriven, and G. Maxwell, eds., Minnesota
Studies in the Philosophy of Science, vol. 2. Concepts, theories, and the mind-body
problem: Minneapolis, Univ. Minnesota Press, 553 pp.

and OPPENHEIM, P., 1948, Studies in the logic of explanation: Phil. Sci., vol. 15,

pp. 135-175.

HOOYKAAS, R., 1959, Natural law and divine miracle: Leiden, E. J. Brill, 237 pp.

HORBERG, L., 1955, Radiocarbon dates and Pleistocene chronological problems in the
Mississippi Valley region: J. Geol., vol. 63, pp. 278-286.

HUTTON, JAMES, 1788, Theory of the earth: Roy. Soc. Edinburgh, Tr., vol. 1, pp. 209-304.

, 1795, Theory of the earth with proofs and illustrations, vol. I: Edinburgh,

Gadell, Junior and Davies, 620 pp.

KAY, M., 1951, North American geosynclines: Geol. Soc. Am., Mem. 48, 143 pp.

KITTS, D. B., 1963, Historical explanation in geology: J. Geol., vol. 71, pp. 297-313.

LEET, L. D., and JUDSON, S., 1954, Physical geology: New York, Prentice-Hall, 466 pp.

LEOPOLD, L. B., and LANGBEIN, W. B., 1962, The concept of entropy in landscape evolu-
tion: U. S. GeoL Survey, Prof. Paper 500-A, pp. A1-A20.

LONGWELL, C. R., and FLINT, R. F., 1955, Introduction to physical geology: New
York, John Wiley, 432 pp.

LYELL, CHARLES, 1881, Life, letters and journals of Sir Charles Lyell Bart, Mrs. Lyell,
ed.: London, John Murray, 489 pp.

MACKIN, J. H., 1938, The origin of Appalachian drainage a reply: Am. J. Sci., vol.
236, pp. 27-53.

, 1948, Concept of the graded river: Geol. Soc. Am., B., vol. 59, pp. 463-512.

MOORE, R. C., 1958, Introduction to historical geology, 2nd ed.: New York, McGraw-
Hill, 656 pp.

NAGEL, E., 1961, The structure of science: New York and Burlingame; Harcourt,
Brace, and World, 618 pp.

READ, H. H., 1957, The granite controversy: New York, Interscience publishers, 430 pp.

SCRIVEN, M., 1959, Truisms as the grounds for historical explanations, pp. 443-475 in
P. Gardiner, ed., Theories of history: Glencoe, Illinois, The Free Press, 549 pp.

TURNER, F. J., 1948, Mineralogical and structural evolution of the metamorphic rocks:
Geol. Soc. Am., Mem. 30, 342 pp.

VON ENGELN, O. D., 1942, Geomorphology: New York, Macmillan, 655 pp.

, and CASTER, K. E., 1952, Geology: New York, McGraw-Hill, 730 pp.

WELLER, J. M., 1960, Stratigraphic principles and practice: New York, Harper, 725 pp.

V. E. McKELVEY

U. S. Geological Survey

Geology As the Study of
Complex Natural Experiments 1

Geology offers almost unique opportunity to observe the results of processes
that not only involve interplay of more variables and larger masses than can
be handled in the laboratory, but that also extend over much greater periods
of time and hence reveal the effects of reactions too slow to observe under
ordinary conditions. Other natural sciences, of course, offer similar oppor-
tunities to observe the interplay of many variables on a grand scale, but none
save astronomy observes the complete record and results of "natural experi-
ments" that have required millions or billions of years to complete.

Because it observes the results of complex natural experiments conducted
on a large scale in both time and space, geology is an exploratory science that
provides opportunity to observe phenomena and processes not predictable on
the basis of knowledge and theory acquired and developed through the lab-
oratory sciences alone. In the past, geology has therefore guided the concepts
and set limits to the scope of the laboratory sciences, and it will continue to do
so in the future. Beyond this, geology provides means of solving a wide variety
of problems too complex to be attacked by artificial experiments.

Like artificial experiments, the experiments of nature are of little value unless
their results are carefully observed and viewed through the frame of searching
question and imaginative hypothesis. However, approached in this manner,
they have led to the discovery of new phenomena and to the formulation of
principles that might not have been developed otherwise. A few examples
will illustrate this point.

1 In recent years, I have discussed the significance of geology as an exploratory and
problem-solving science with a number of friends whose ideas I am sure are reflected
in this essay. I wish especially to acknowledge stimulating discussions on the subject
with J. R. Balsley, W. H. Bradley, R. M. Garrels, R. A. Gulbrandsen, and H. L. James,
and to thank for similar stimulation as well as specific suggestions on this manuscript
C. A. Anderson, Michael Fleischer, E. D. Jackson and E. M. Shoemaker.

69

70 V. E. McKELVEY

Discovery of the Nonpredictable

Observations of natural phenomena lie at the root of all science, and most of
these phenomena were at some stage not predictable on the basis of accumu-
lated human knowledge and theory. Perhaps more than other realms accessible
to human observation, the earth has been the source of the discovery of phe-
nomena that have stimulated scientific inquiry or posed challenging problems
which led eventually to the principles forming the web of physical science.
Thus observations of minerals posed many of the problems that stimulated the
growth of chemistry, crystal chemistry, crystallography, and optics; observa-
tions of the properties of uranium minerals and compounds formed the base
of both radiochemistry and nuclear physics; and the discovery of the natural
phenomena of fluorescence, luminescence, and magnetism played similar roles
in other branches of physics. At the time these various phenomena were ob-
served they could not have been predicted from available knowledge and
theory, and indeed many of them would be unpredictable even today if they
were not known to exist in nature.

There would not be much point here in calling attention to the study of the
earth, its processes, and history as a source of discovery of phenomena stimu-
lating general science, if this form of contribution were a thing of the past.
The fact is, however, that geology is still playing an important role as an ex-
ploratory science and has abundant opportunity to do so in the future. Some
of these discoveries may come in the form of single observations for example,
T. F. W. Barth's and E. Posnjak's discovery of the inverse structure of the
spinels, which contributed significantly to the theory and development of semi-
conductors; and K. J. Murata's discovery that the fluorescence of natural
halite and calcite, previously ascribed to the presence of manganese, actually
occurs only when two activating elements, lead and manganese, are present, a
phenomenon that led to the synthesis of many new fluorescent compounds.
Some of the most significant discoveries, however, are likely to be of complex
phenomena, including those of a secular nature, whose recognition depends on
the collection and synthesis of many and perhaps diverse data. The concepts
of biologic evolution, folded geosynclines, and periodic reversal in polarity of
the earth's field are examples of secular processes discovered through the
synthesis and analysis of many diverse geologic observations.

The process of piecing together and analyzing isolated geologic observations
is notoriously difficult and slow, not only because of the number and diversity
of the phenomena involved but also because of their scale. Thousands of
square miles may have to be mapped, scores of stratigraphic sections measured
and correlated, volumes of data accumulated on rock properties, and many
hypotheses tested and discarded before even a valid empirical relation among
such data can be formulated.

The fable of the blind men and the elephant the incorrect conclusions
reached by observers whose vision is limited compared to the breadth and

THE STUDY OF COMPLEX NATURAL EXPERIMENTS 71

complexity of the problem has sometimes been cited to ridicule the stumbling
operation of inductive reasoning. This tale, however, carries a meaning of far
deeper significance than is generally recognized in the telling, namely, that it
is only by patient exploration that an elephant or a folded geosyncline can
be discovered. Once the existence of an elephant has been established, many of
its other characteristics can be deduced, but accumulated scientific theory is
not yet and may never be adequate to predict its existence in the first place.

What are some of the areas in which geologic elephants are likely to be dis-
covered in the future? Because I am talking about phenomena that are not
predictable, I can hardly be expected to be very specific, but a few examples
of some of the poorly explored areas that may yield the unexpected are in
order. Surely the subject of crustal structure and composition is one such prom-
ising field; the relation between local structure, thickness, and composition of
rocks of the upper crust, and similar properties in the lower crust and upper
mantle are as yet poorly known and can be expected to yield relationships not
imagined now. In stratigraphy and paleontology, data adequate to permit a
comprehensive view of the paleogeographic, paleoclimatic, paleoecologic, and
paleotectonic records are just now being synthesized, and their analysis very
likely will disclose relations not now anticipated: between rate and direction
of biologic evolution and environmental controls, between geologic history and
secular changes (perhaps related to unidirectional or evolutionary changes in
the composition of the ocean and the atmosphere) in the composition of sedi-
mentary rocks, and between tectonic history and climatic changes. Data on
the abundance of the elements, especially the minor ones, and on the mass of
major rock units in the earth's crust are as yet so sparse and poorly integrated
that we can still expect surprises on the subject of the composition of the earth's
crust and its major components. Generally speaking, the data being accumu-
lated on rock structure the wavelength and amplitude of folds; the distribu-
tion, orientation, and frequency of fractures, the magnitude of the displace-
ments along them, and the orientation of minerals within them have not
been synthesized and analyzed with respect to such factors as area! magnitude,
thickness, and composition of the deformed units, depth of burial or load, and
duration of the deforming forces. Finally, integration of knowledge in all these
areas will be necessary to test such concepts as continental drift and an expand-
ing or shrinking earth. No doubt other elephants will be discovered in the
process.

Use of Natural Experiments in Solving Complex Problems

In using nature's completed experiments to solve problems involving such a
large number of variables, large mass, or long period of time that they cannot
be attacked satisfactorily in the laboratory, we begin with the problem or
question and observe results of experiments already performed. What minerals

72 V. E. McKELVEY

will precipitate from a complex silicate melt, given time for slow reactions to
reach equilibrium? Igneous differentiates and their ores will yield the answer
in a detail unobtainable otherwise. Can specific minerals ordinarily formed at
high temperatures form at low temperatures, given time? Their presence or
absence as authigenic minerals in sedimentary rocks may suggest the answer
(and incidentally Charles Milton's work on authigenic minerals formed in
saline deposits has shown that the list not only includes such well-known forms
as feldspar but also many other high-temperature forms, such as riebeckite
and leucosphenite). What diadochous substitutions, including doubly com-
pensated ones, can occur in a given compound? Analysis of mineral specimens
from a variety of geologic environments is likely to yield a more comprehensive
answer than artificial experiments do. What are the relative solubilities of
refractory minerals in dilute aqueous solutions? Their behavior during weath-
ering will give valuable qualitative data, and if analyzed with regard to the
characteristics of soil solutions, runoff, and mean annual temperature, it may
be possible to obtain more quantitative answers. To what extent will brittle
minerals deform plastically, and under what physical conditions? Natural
experiments provide a parameter time that cannot be evaluated in the
laboratory.

The usefulness of natural experiments in problem solving depends, of course,
on the degree to which the conditions of the experiment can be established,
and this may require use of both field and laboratory data. For example, the
relative conditions under which syngenetic minerals are deposited from sea-
water can be established from lithofacies analysis of beds known from geologic
evidence to have been synchronously deposited. If physical and chemical data
are available for a few of the minerals, however, the conditions under which
the others have been deposited may be interpolated. Using this approach,
H. L. James demonstrated from stratigraphic analysis that iron sulfides, iron
silicates, iron carbonate, and the iron oxides, hematite and (most surprising,
for it is ordinarily thought of as a high-temperature mineral) magnetite, are
synchronously deposited in a regular lateral sequence reflecting progressive
change in redox potential and pH. And R. P. Sheldon, my co-worker in the
study of the phosphorite family of sediments, defined rather closely, from litho-
facies analysis and existing physical and chemical data, the physical and
chemical conditions in the Phosphoria sea, in which black shale, phosphorite,
chert, carbonates, redbeds and salines were deposited in a laterally changing
environment ranging from the deeper part of the shelf to coastal lagoons.

The work of E. M. Shoemaker and others on meteorite craters illustrates the
deliberate use of geology in solving problems of multidisciplinary interest.
Measurements of the depth and radius of craters formed artificially by chemical
and nuclear explosives and the radius of mixing due to shock show a consistent
relation to yield, and similar features in meteorite craters appear to obey the

THE STUDY OF COMPLEX NATURAL EXPERIMENTS 73

same scaling laws. The dimensions and other features of natural meteorite
craters are equivalent to those of craters formed by the impact of explosions
in the range of tens or even hundreds of megatons. Study of their structure
and other features, therefore, yields data valuable in predicting the effects of
nuclear explosives that are more powerful than those tested artificially in the
past, as well as the effects of nuclear explosions in rock types, such as granite
and limestone, in which nuclear devices have not yet been tested. Artificial
tests thus far have been limited to a few rock types, and provide little basis for
predicting the effects of natural structures, such as joints, on crater shape. A
natural experiment, Meteor Crater, Arizona, already completed in jointed
limestone, however, shows that joints do influence crater shape, for the outline
of Meteor Crater is nearly square and the diagonals of the square correspond
to the direction of the two principal joint systems. It is interesting to note also
that the discoveries of the high-pressure silica polymorphs, coesite and stisho-
vite, by E. C. T. Chao and co-workers, in impactites associated with meteorite
craters demonstrated for the first time that these high-pressure phase changes
could be produced by shock pressure of extremely short duration.

The power of laboratory experiments stems in large part not only from the
fact that they can be closely controlled, but also that they can be repeated to
isolate and measure the effect of a single variable. This advantage is not
wholly lost in natural experiments, however, for similar experiments may have
taken place under differing circumstances that bring to light the effects of
various factors. For example, according to E. D.Jackson (personal communica-
tion, 1963), the composition of the original magmatic liquid in the Precambrian
Stillwater complex of Montana and the still liquid lava lake at Kilauea Iki,
Hawaii, is essentially the same, so that differences in their crystallization history
and products can be related to their size and position in the crust.

Summary

Geology, like other sciences, has many economic or practical objectives that
in themselves more than justify its pursuit and support. The knowledge and
principles it develops enable us to use more fully the land, water, mineral, and
energy resources of our physical environment and if it yielded no other
dividends it would, for these reasons alone, deserve high emphasis among scien-
tific endeavors in the modern world. The opportunities it provides, however,
for observing the results of natural experiments make it important also as one
of the prime exploratory and analytical fields of general science, capable not
only of uncovering challenging problems but also of solving many difficult and
complex ones that are beyond the reach of artificial experiments.

74 V. E. McKELVEY

REFERENCES CITED

EARTH, T. F. W., and POSNJAK, E., 1931, The spinel structure: an example of variate

atom equipoints: J. Washington Acad. Sci., vol. 21, pp. 255-258.
CHAD, E. C. T., FAHEY, J. J., and LITTLER, J., 1962, Stishovite, SiC>2 5 a very high pres-
sure new mineral from Meteor Crater, Arizona: J. Geophys. Res., vol. 67, pp. 419-421.
, SHOEMAKER, E. M., and MADSEN, B. M., 1960, First natural occurrence of

coesite: Science, vol. 132, pp. 220-222.
JAMES, H. L., 1954, Sedimentary facies of iron-formation: Econ. Geol., vol. 49, pp.

235-293.
MILTON, CHARLES, 1957, Authigenic minerals of the Green River formation of the

Uinta Basin, Utah: Intermountain Assoc. Petroleum Geol., 8th Ann. Field Conf.,

pp. 136-143.
MURATA, K. J., and SMITH, R. L., 1946, Manganese and lead as coactivators of red

fluorescence in halite: Am. Mineralogist, vol. 31, pp. 527-538.
SHELDON, R. P., in press, Physical stratigraphy and mineral resources of Permian rocks

in western Wyoming: U. S. Geol. Survey, Prof. Paper 313-B.
SHOEMAKER, E. M., 1960, Brecciation and mixing of rock by strong shock: U. S. Geol.

Survey, Prof. Paper 400-B, pp. 423-425.
, 1960, Penetration mechanics of high velocity meteorites illustrated by Meteor

Crater, Arizona: Internat. Geol. Cong., 21st, Copenhagen, 1960, pt. 18, sec. 18,

Pr., pp. 41 8-434.

A. O. WOODFORD

Pomona College

Correlation by Fossils

The presence of successive and dissimilar fossil faunas in the stratified rocks
of northwestern Europe was demonstrated by William Smith and his con-
temporaries as early as 1815. Some forty-five years later, Darwin convinced
the scientific community that evolution of stratigraphically lower faunas into
higher ones is more probable than alternating creations and extinctions. Soon
after the "Origin of Species" appeared, however, Thomas Henry Huxley
(1862, 1870) challenged the assumption, already well established in his time,
that two widely separated sedimentary rock masses containing closely similar
faunas or floras must have been deposited at the same time. He asked for a
sharp distinction between homotaxis (identical or similar successions of faunas
or floras) and the kind of correlation that implies synchronism (identical age for
each correlated faunal or floral pair). Huxley (1862, p. xlvi) asserted that
"a Devonian fauna and flora in the British Islands may have been contempo-
raneous with Silurian life in North America, and with a Carboniferous fauna
and flora in Africa."

I have no intention of championing this heresy, but I do intend to follow
Huxley's logic, working from homotaxis to possible time correlation, without
making the assumption that homotaxis necessarily implies synchronism.

Guide Fossils

Correlation by fossils involves two problems. First we must determine the
changes in the successive fossil faunas at one locality; and second, we must
learn how to use the faunal changes in making correlations from one local
column to another, carefully distinguishing between time and facies correla-
tions. Even while working on a single column it is desirable to keep two
objectives in mind. Paleontologically, we use our collections of fossil faunas
relatively small and probably biased samples of the original, much larger,
biological populations as bases for unscrambling the ecological facies and
evolutionary sequences of the animals. Stratigraphically, we subdivide the

75

76 A. O. WOODFORD

column by use of differences between the species at different levels. Note that
the evolutionary and the stratigraphic values of species require emphasis on
exactly opposite aspects of the observations. For evolution we emphasize
similarities between the species at successive levels, and for correlation we
emphasize differences.

Subdivision of the column depends on how fossiliferous the rocks are and
how clearly the succession is exposed. If the local succession is well exposed
and the faunas are rich, we must choose between two possible methods of
work: (1) statistical study of a large collection of each fauna (or a large sample
of the ammonites or brachiopods or other selected group), a procedure that in
the past has usually seemed excessively time-consuming, or (2) use of selected
guide species (or genera), the customary procedure. With computers at hand
to make the calculations, statistical methods are coming into vogue, but for
any general discussion it is still necessary to stick to guide fossils.

The ideal set of guides would belong to a group that evolved rapidly, occurs
abundantly in all kinds of stratified rocks, and is worldwide in distribution.
The fossils that approach this ideal most closely appear to be the Ordovician-
Silurian graptolites, the Upper Paleozoic fusuline foraminifera, the Upper
Paleozoic goniatitic ammonoids, the ammonites of the three Mesozoic systems,
and a few Cretaceous and Tertiary planktonic foraminifera, and still smaller
nannoplankton. The ammonites appear to be the best of the lot. Among the
Mesozoic systems, the Jurassic includes, in northwestern Europe, the most
usefully fossiliferous sequence, with excellent preservation and the most favor-
able set of well-studied exposures. Almost every aspect of stratigraphic sub-
division by the use of guide fossils can be illustrated from the Jurassic System,
in Europe and elsewhere.

Lower and Middle Jurassic Zones of Northwestern Europe

Subdivision of Jurassic. The Jurassic has been divided and subdivided into
stages, substages, zones, and subzones. We shall take the stratigraphic zone as
our starting point. It is true that zones were not the first divisions of the Jurassic
to be invented. Historically, several of William Smith's formations Lias,
Blue Marl, Under Oolyte, Great Oolyte, Forest Marble, Cornbrash Lime-
stone, Kelloways Stone, Chinch Clay, Coral Rag, Oak Tree Clay, Portland
Rock, and Purbeck Stone were put together to form the Jurassic System.
Then Alcide d'Orbigny (1842-1851; summary, pp. 600-623) reversed the
process and divided most of the Jurassic into 10 paleontologically defined
stages, using ammonites primarily. Finally, the stages were subdivided into
zones by d'Orbigny and others, notably A. Oppel of Germany (1856-58).
Current usage for the Lower and Middle Jurassic is shown in Table 1 on
pp. 78-79.

CORRELATION BY FOSSILS 77

Zones. l A zone is made up of the stratified rocks deposited with a particular
assemblage of fossils. This definition is vague and very nearly begs the vital
time question, but it has nevertheless proved fairly easy to use. \Ve shall come
back later to the possible time significance of a zone.

The implications of the zonal concept are grasped most readily in connection
with an idealized example. First, we study in detail a sequence of beds at one
location, A. We study and name the fossils contained therein, emphasizing the
ammonites in the Jurassic. Almost always the ammonite succession is dis-
continuous; the section yields several distinct successive faunas. We choose
one species in each fauna as a special guide or index and use its name to label
the fauna. The limits of the faunal ranges do not necessarily coincide with
rock boundaries.

We repeat the process at a second exposure, B. The rock succession is differ-
ent, but the faunal succession at B has elements in common with that at A.
In particular, some of the faunal zones established at A can be recognized at 5,
even though others cannot. The reverse may also be true. In this manner we
compare faunal successions in ever widening geographical extent. We find
that some of our faunas at A can be identified over distances of hundreds of
miles while others are very local. By compilation of successive faunas of suit-
ably wide extent we can construct a table of standard zones for a particular
area, such as northwestern Europe. The names of the standard zones are de-
rived from faunal index species, used to designate stratigraphical sets of beds,
not paleontological entities. A particular zone is referred to as the "Zone of
Parkinsonia parhnsoni" or merely "Parkinsoni Zone" (No. 30 in Table 1).

Now, returning to our starting point A, we see that some of our original
local zones at that place have become standard zones. Others, that proved to
be more or less local in extent, are usually attached to one or another of the
standard zones, as subzones. Moreover, we may find that not all of the standard
zones are represented at A. That is, the section at A is not paleontologically
complete; it contains paleontological lacunas. Finally, of the nonfossiliferous
beds between zones we can say nothing.

Note that (a) for the recognition of a zone, we need both rock and fossils;
(b) a zone is characterized by an assemblage of species (especially that at the
type locality), and the zone index species may be rare or even missing from
any one place or even from a region; (c) it is a matter of convenience, based
usually on geographical extent, as to which units we call zones and which sub-
zones; (d) the existence of a standard zone table implies widespread homotaxis;
and (e) such homotaxis may prove to be explainable only in terms of syn-
chronism (time correlation), making the standard zonal table also a table of

1 The first six paragraphs of this discussion of the zone are modified from a statement
furnished by Dr. J. H. Callomon of University College, London.

TABLE 1

CORRELATION TABLE FOR MARINE LOWER AND MIDDLE JURASSIC,

Stage

Zone

1
England

2

Eastern
France

3
South-
western
Germany

4
North-
western
Germany

Middle

41.

Quenstedtoceras lamberti

z

z

z

z

Jurassic

40.

Peltoceras athleta

z

z

z

z

Callovian

M

39.
38.

Erymnoceras coronation
Kosmoceras jason

z
z

z

z
z

z
z

37.

Sigaloceras callomense

z

z

?

z

36.

Macrocephalites macrocephalus

z

z

z

z

35.

Clydoniceras discus

z

z

z

34.

Oppelia aspidoides

z

p

?

p

Bathonian

M

33.
32.

Tulites subcontracts
Gracilisphinctes progracilis

z
z

z

z

L 31.

Zigzagiceras zigzag

z

z

z

z

30.

Parkinsonia parkinsom

z

z

z

z

U

29.

Garantiana. garantiana

z

z

z

z

28.

Strenoceras subfurcatum

z

z

z

z

27.

Stephanoceras humphriesianum

z

z

z

z

Bajocian

M

26.

Otoites sauzei

z

z

z

z

j w~.

25.

Sonninia sowerbyi

z

z

z

z

24.

Graphoceras concavum

z

z

z

z

T

23.

Ludwigia murchisonae

z

z

z

z

ij

22.

Tmetoceras scissum

z

z

z

21.

Leioceras opalinum

z

z

z

z

Lower Jurassic

20.

Dumortieria levesquei

z

z

z

z

(lias)

U

19.

Grammoceras thouarsense

z

z

z

?

Toarcian

18.
17.

Haugia vanabilis
Hildoceras bifrons

z
z

z
z

z
z

z
z

L

16.

Harpoceras falcijer

z

z

z

z

15.

Dactyhoceras tenuicostatum

z

z

z

z

u

14.

Pleuroceras spinatum

z

z

z

z

\J

13.

Amaltheus margaritatus

z

z

z

z

Pliensbachian

12.

Prodactylioceras davoei

z

z

z

z

L

11.

Tragophylloceras ibex

z

z

z

z

10.

Uptonia jamesoni

z

z

z

z

9.

Echioceras raricostatum

z

z

z

z

U

8.

Oxynoticeras oxynotum

z

z

z

z

Sinemurian

7.
6.

Asteroceras obtusum
Caenisites turnen

z

z

z
z

z
z

z
z

L

5.

Arnioceras semicostatum

z

z

z

z

4.

Anetites bucklandi

z

z

z

z

3.

Schlotheimia angulata

z

z

z

z

Hettangian

2.

Alsatites hasicus

z

z

z

z

1.

Psiloceras planorbis

z

z

z

z

t Compiled from W. J. Arkell's Jurassic Geology of the World, with modifications from
Dean, Donovan, and Howarth (1961), J. H. Callomon (1955, 1961), and H. K. Erben (1956).
U = Upper. M = Middle. L = Lower. S = stage present, z = zone present, x = fauna

78

SHOWING ZONES AND STAGES IN 13 SELECTED COLUMNS!

5

Jura
Mts.

678 9 10 11 12 13

Cau- A . . Cutch, Eastern Western ,., . Eastern A ,
casus Arabia India Greenland Canada ^ onun S Mexico Andes

z

ss z z

41

z

z z

40

ss

ss Z J + S

39
38

z

z ? z + X
ss

37

z

z z z +

36

z

ee? OO

35

z

aoi oo ^

34

z

S 1

33

p

SS w

32

z

? 1 ? ? z

31

z

z

30

z

SS SS?

29

z

Z

28

z

z z z

27

z

z ss z ss S ?

26

z

? z z

25

z

?

24

z

z

23

z

z z

22

z

z

21

?

20

z

ss? ss ss ss

19

*

s

18

*

z

17

*

ss

16

*

ss ss

15

z

14

z

ss

13

?

S z S z

12

z

z

11

z

z z z

10

z

z

9

z

? ss

8

z

s? ? ? ?

7
6

z

z z ?

5

z

z z z

4

z

z ?

3

z

s?

2

z

z z

1

bridging Middle and Lower Callovian. ? = zone perhaps present. 1 = 6 local Bathonian
zones. + = 5 local zones, ss substage present, ss? = substage perhaps present. * = per-
haps all zones; condensed.

79

TABLE 2

DISTRIBUTION OF AMMONITE SPECIMENS IN BETHEL BRICK QUARRY AT

Zone

29-30
transition

29 (G. garantiana), 13 meters

Bed number

1

2 3-4 7 8 15 16-17 19 21

Parkinsonia acns and P. rancostata
Garantiana (Subgarantiana) depressa

6
2

Garantiana (Subgarantiana) tetragona
Garantiana (Subgarantiana) suemca
Garantiana (Subgarantiana) alticosta
Garantiana (Subgarantiana) subgaranti
Garantiana (Subgarantiana) wet&li
Garantiana (Subgarantiana) trauthi
Garantiana (Subgarantiana) pompeckji
Garantiana (Subgarantiana) coronata
Garantiana (Subgarantiana) cydogaster
Garantiana (Subgarantiana) subangulata
Garantiana (Garantiana) garantiana
Garantiana (Garantiana) dubia
Garantiana (Garantiana) baculata
Garantiana (Garantiana) althoffi
Garantiana (Garantiana) filicosta
Garantiana (Psmdogarantiana) minima
Garantiana (Pseudogarantiana) dichotoma
Perisphinctids
Bigotites sp. indet.
Garantiana (Qrthogarantiana) mflata
Garantiana (Orthogarantiana) schroederi
Garantiana (Orthogarantiana) densicostata
Strenoceras subfurcatum
Strenoceras bajocensis
Strenoceras latidorsatum
Strenoceras robustum
Strenoceras rotmdum

164361

1 1

2

time correlation, with each zone made up of the rocks deposited during the
time when the group of zonal guide fossils was living.

Historically, Oppel set up the first table of standard Jurassic zones for north-
western Europe in 1856. This event was fortunately located, for the unusually
complete and extensively fossiliferous European sequence gives geographic
zone ranges of the order of 1000 miles, and the experience of the hundred years
after Oppel has shown that this is a suitable extent for a zone.

80

BIELEFELD, NORTHWEST GERMANY. AFTER A. BENTZ, 1928.

28-29 transition, 12 meters . 28 (S ~ ^/

1 4 meters

22 23 24 25 26 27 27A* 28 29-32 34 35 36 41 42 , Upper Lower

1 1

1

142 1

2" 1 64

3

2 1
1

1 l a 1

1

1

1

1 1 1

1 4 1

4 4

3 1

2
1
1 2 8

2

2

15

5

11

2?

4

1

2

2

* Specimens from nodules, not all from one horizon.
*cf.

Some peculiarities of actual faunas are shown in Table 2, representing the
ammonites of a sequence of Upper Bajocian units in the clays and shales of a
brick quarry near Bielefeld in northwestern Germany. The Garantiana Zone
(No. 29 of Table 1) and the Subfurcatum Zone (No. 28) are present and in
addition, there are the 28-29 and 29-30 transitional units. Here the paleon-
tological units are also lithologic units. The 29-30 transitional beds, at the top

81

82 A. O. WOODFORD

of the section, are blue-black micaceous shales. Zone 29 is made up of 13
meters of more calcareous shales with few fossils, though some ammonites are
present, as shown in the table, and a reptile skeleton was found at the base of
the zone. The 28-29 transition beds, which are 12 meters thick, are somewhat
micaceous marly clays. Fossils are moderately numerous. Zone 28, that of
Strenoceras subfurcatum, is four meters thick and is distinguished by its fauna.
Paleontologically, the transition beds between zone 30, that of Parkinsonia
parkmsom, and zone 29, that of Garantiana garantiana, are characterized by the
occurrence together of Parkinsonia and Garantiana. Zone 29, at this locality,
yielded but a single example of the index species, Garantiana garantiana, and is
characterized by perisphinctids. The 28-29 transition beds have the distinction
of being the only ones containing Pseudogarantiana. In zone 28 the zone species
Strenoceras subfurcatum is common; in this section, the genus Strenoceras is limited
to this zone.

At Bielefeld, paleontological subdivision goes beyond the standard zones in
two ways, (a) Two extra units are present, both well marked paleontologically,
and these units contain faunas which are transitional from one standard zone
to the next. Such transitional units are unusual in the Jurassic of northwestern
Europe, (b) The Subfurcatum Zone is divided into two subzones. Subzones
are more numerous in some other parts of the northwest European Jurassic.
The 20 zones of the Lower Jurassic (Lias) include 50 subzones, most of which
have been recognized in most parts of the province (Dean and others, 1961).

Kosmoceras in the Callovian at Peterborough. For details concerning species and
specimens we turn to a part of the Callovian, here considered the uppermost
stage of the Middle Jurassic, following the practice of Arkell (1956) and the
German geologists. Some of the commonest Callovian ammonites exhibit
nearly continuous transitions from one species to another, and some of the
vaguely delimited species are used as guide or even index fossils. With the aid
of these guides, exceptionally satisfactory homotaxial parallels have been
demonstrated, notably between the sections in the Oxford (Clunch) Clay at
Peterborough and Kidlington, England.

The Peterborough section was examined by Roland Brinkmann of Germany
(1929) in a statistical study that involved 3035 specimens which he assigned to
the genus Kosmoceras. Brinkmann also identified the few other ammonites he
found, as well as the numerous pelecypods, belemnites, and other invertebrates
that he collected. He mentioned the fine fish specimens found in the quarries
by others, as well as the skeletons of large plesiosaurs that are exhibited in the
British Museum.

The Oxford Clay near Peterborough is flat-lying and at least 98 ft thick.
It is mostly a dark gray, poorly laminated bituminous rock. The 1300 cm
(42 ft) of clay in the lower part of the formation, which was studied statistically

CORRELATION BY FOSSILS 83

by Brinkmann (compare Table 3), is exposed in the lower parts of the walls
of three quarries of the London Brick Company. One of the quarries is two
miles south of Peterborough; the others are close together, three or four miles
southeast of the city. The clay contains numerous white layers called plasters,
usually only a few millimeters thick, rarely several centimeters, which separate
the dark-colored rock into bands. The plasters are made up of collapsed am-
monite shells, mostly in separated fragments. Local lithologic correlation,
from one part of a pit to another part, or between pits, is made possible by the
plasters and by three much thicker reference beds: (1) a concretionary* layer
56-78 cm above Brinkmann's zero horizon (perhaps 176-198 cm above the
Kellaways Rock that underlies the clay), (2) a very dark, thick-bedded clay
at 560-680 cm above the zero level, and (3) a green clay at 1130-1160 cm.
The thicknesses of the plasters and other reference beds and the intervals
between them are uniform throughout the Peterborough quarries.

Brinkmann recognized 12 Kosmoceras species at Peterborough in the 1300 cm
of clay to which the statistical study was restricted (Fig. 1). He concluded that
each of the four columns of species in Fig. 1 represents an evolutionary lineage
that should be recognized as a subgenus of Kosmoceras. Each subgenus is desig-
nated by one of the many names previously coined by S. S. Buckman: %ugo-
kosmokeras (column 1), Kosmoceras s. s. (column 2), Gulielmiceras [Anakosmokeras]
(column 3), and Spimkosmokeras (column 4). Other workers, notably Callomon
(1955), separate the part of the first column below the 135.5-cm horizon
(K. jason and lower) as the subgenus Guhelmites (another of Buckman's names).

The shells of the last two columns are similar to those of the first two, horizon
by horizon, except for smaller size and the possession of prongs called lappets
on either side of the aperture (the back lappet of a pair being hidden by the
front one in a side view). At almost ever}* horizon a form figured in column 3
corresponds to one of similar sculpture in column 1 that is twice as big and
lacks lappets. The same relation holds somewhat less clearly between columns
2 and 4. Brinkmann (1929, pp. 212-213) concluded that the dimorphism is
not sexual and that the unlappeted lineages were derived from Sigaloceras
(Table 4) slightly before the beginning of Oxford Clay deposition. Now J. H.
Callomon (letter, February 10, 1963) doubts the distinctness of the four lineages
and considers the Kosmoceras sequence at Peterborough a single lineage, with
sexual dimorphism. Brinkmann's collections actually were not large enough
to demonstrate bimodal distributions.

Whatever the number of lineages, Brinkmann's statistics are useful, in part
because of the numerical predominance of ugokosmokeras (column 1 of Fig. 1),
with 1802 specimens, 703 complete, over Kosmoceras s.s. (column 2), with 67
specimens, 50 complete. The statistics are consistent with an explanation in
terms of persistent evolutionary changes, if there are lacunas in the fossil
record at some of the plasters. Brinkmann grouped measurements and ratios

84

Grossouvrei

Grossouvrei

obductum

obductum

Jason

enodatum

gemmatum

A. O. WOODFORD
Section height, cm

-1300

-900

pollucinum

-800

-700

-100

nodosum

Fig. 1. Lineages and subgenera of Kosmoceras at Peterborough, modified slightly
from Brinkmann (1929, Tafel V). 1, ugokosmokeras and (Jason and lower) Gvliclmites;
2, Kosmoceras s.s. Sizes greatly reduced.

CORRELATION BY FOSSILS

85

compression

Section height, cm

aculeatum om ^

Coarse ribbed Typical Fine ribbed

Castor

Pollux

1300

-900

800

700

100

Fig. 1 (Cont.}. 3, Gulielmiceras; 4, Spinikomokeras. Sizes greatly reduced.

86 A. O. WOODFORD

by lineages and horizons, the 1300 cm of clay being divided into 48 strata
groups. The measurements for one lineage in one strata group tended toward
normal distribution. Two of the many sets of data are selected for presentation
here. The first set is shown as Table 3, representing the variations in diameter
among the 703 complete specimens, most of them flattened to thin discs, that
were assigned to the Gulielmites-^ugokosmokeras lineage (column 1 of Fig. 1).
Note the rapid increase in size up to the 135-cm plaster, the sharp drop there,
and the slow and interrupted increase thereafter, with the maximum mean
diameter 146.6 2.0 mm for the 31 specimens from 793 cm (individuals up to
almost 170 mm). Standard deviations of the differences between adjacent
means have been added to the table, together with /-values (difference between
adjacent means divided by the standard deviation of the difference). All
/-values greater than 3 for the differences 1, 3, 4, 6, 10, and 27 are strongly
indicative of dissimilar populations. Difference 10, at the 135.5-cm horizon, is
between the highest Jason Zone strata group and the lowest Goronatum Zone
straca group. These two zones are the two parts of the Middle Callovian
(compare Table 4). Most other pertinent statistical measures for all lineages
also show breaks at this horizon.

The second set of Brinkmann statistics pertains to the ratio between the num-
bers of outer ribs and peripheral spines in the Spinikosmokeras sequence of Fig. 2.
In this sequence, note the change from the approximately 1/1 ratio between
outer ribs and spines in Fig. 2(c) to the approximately 2/1 ratio in Fig. 2(d).
The statistics (Fig. 3a) show that this change is largely concentrated between
1093 and 1094 cm (1093.5 cm of Fig. 3a), at a horizon which is now about to
be established as the Athleta-Coronatum zonal contact (J. H. Callomon, letter
of February 10, 1963; see also Table 4). The regression lines on the two sides

Fig. 2. A Spinikosmokeras succession at Peterborough, after Brinkmann (1929, Tafel
III), (a) K. (S.) castor anterior, from 312 cm; (b) JT. () castor castor, from 670 cm; (c)
K. (S.) aadeatum anterior, from 988 cm; (d) K. (S.) aculeatum acvleatum, from 1277 cm.
Approx. X J natural size.

CORRELATION BY FOSSILS

87

3.0-
1 25^

.

1 ' :

M

,'

1 2.0-

. "

I 1 - 5 -

. .; -

-^r^T^-

i n-

^H^*

1080

1100

1120 1140

1093.5
Centimeters above zero level

(a)

3.0

|2.5-

Lacuna

1&2.0-

1 1-5-

1 O

:

^^ ~-

.. :. ./. -

i i i ,

1 /-kOrt

i i i i i i i

1 1 f\f\ 1 1 Ort 1

1093.5

1093.5 '

CentI meters
(b)

Fig. 3. Lacuna at Peterborough, shown by Spinikomokeras statistics and regression
lines. After Brinkmann (1929, Abb. 20). (a) plotted by strata-groups; (b) expanded
to evaluate lacuna.

of the 1093.5-cm plaster actually have the same slope. This slope represents a
rate of change that Brinkmann attempted to use in estimating the relative
length of time represented by the lacuna. By assuming that (1) the rib-spine
ratio changed at a uniform rate, (2) the clay between the plasters accumulated
at a uniform rate, and (3) each plaster represents a time of nondeposition of
clay, he was enabled to draw a second diagram with a gap at 1093.5 cm, wide
enough to produce a single straight rib-spine regression line (Fig. 3b), in which
the new abscissas became measures of time and from which the relative lengths
of the two periods of clay deposition and the intervening lacuna could be
determined. The result is dependent on the slopes of the two partial regression
lines of Fig. 3(a), which are not very precise because of the scatter of the rib-
spine ratios. Despite this and the other uncertainties, Brinkmann's data show
clearly enough that some plasters, both interzonal and intrazonal, are horizons
of exceptionally large changes in the statistical properties of the ammonites

TABLE 3

END-DIAMETERS OF ZUGOKOSMOKERAS (AND GULIELMITES) FROM

Strata-group,

Number of

Mean and its

For

the differences:

cm

specimens

Stand. Dev., mm

Number

Stand. Dev.

t

7-20

12

61.6 1.5

219

in nn

21-25

12

82.8 1.5

.l

iU.UU

9

9 49

OQ1

26-28

23

85.0 =fc 1.9

Atm'Tft

. y L

i

2in

1 94

29-39

32

78.2 =h 0.9

. 1U

J,t*T

A

1 A

A. f\A.

40-45

16

84.9 d= 1.4

1 .00

^.L/*r

5

3.67

0.98

46-50

9

88.5 d= 3.4

^ n^

^ AA.

56-78

19

105.8 =h 3.7

j.\jj

J.*r*r

A QA.

079

79-90

8

109.3 d= 3.1

*r.O"r

. 1 ft

8

3.80

1.42

91-120

15

114.7 =fc 2.2

9

3.72

1.16

121-135

9

119.0 3.0

in

379

690

136-160

9

95.6 d= 2.2

. /

.*y

11

3.25

0.22

161-200

11

96.3 =b 2.4

12

3.76

0.40

201-240

5

94.8 db 2.9

13

3.58

0.31

241-260

12

93.7 =b 2.1

14

3.83

0.52

261-300

7

95.7 d= 3.2

15

5.12

2.05

301-320

5

85.2 =b 4.0

16

4.47

1.30

321-340

14

91.0 d= 2.0

17

4.20

1.64

341-360

7

97.9 3.7

18

4.25

0.99

361-380

7

93.7 =t 2.1

19

3.50

2.37

381-440

6

102.0 =fc 2.8

20

4.18

0.53

441-460

12

99.8 d= 3.1

21

5.46

2.45

461-500

6

113.2 =fc 4.5

22

5.76

0.40

501-520

19

110.9 =h 3.6

23

4.11

2.21

* Not a plaster; no data; ammonites not classified because of lack of time.

t First 3 columns from Brinkmann, 1929, p. 103. Plasters (and strata-groups whose
fossils were not studied) shown by horizontal lines. Stand. Dev. Standard Deviation.
For t see text.

88

QUARRIES NEAR PETERBOROUGH, ENGLANDJ

Strata-group,
cm

Number of
specimens

Mean and its
Stand. Dev., mm

For

Number

the differences:
Stand. Dev.

t

521-530
531-539

20
13

101.8 2.0
106.1 =h 2.5

24

3.20
3.14
3.14
2.92
3.80
4.14
4.91
5.50
5.58
4.93
10.05
10.05
3.69
3.98
9.53
11.19
7.39
4.76
3.84
3.61
3.48
4.56
4.94
4.32

1.34
2.13
0.03
5.24
0.34
0.89
2.59
2.47
2.29
0.51
1.73
0.01
1.17
0.58
1.45
0.76
2.07
1.43
2.79
1.16
1.72
1.47
0.18
0.60

540
541-559

38
10

112.8 =t 1.9
112.7 d= 2.5

26

07

560
561-680
681-690

27
15
13

128.0 1.5
126.7 =t 3.5
130.4 2.2

28
29
*^n

691-759

13

117.7 =h 4.4

^1

760-780

?

16

131.3 d= 3.3

Jt

u u ^O

781-792
793

8
31

144.1 =Jr 4.5
146.6 d= 2.0

jZ

33

\A

794-854

6

129.2 =fc 10.3

^

855
856-864

23
21

129.1 d= 2.0
124.8 =h 3.1

36

^7

865
866-880
881-894
895

33

7
7
9

127.1 =fc 2.5
140.9 =h 9.2
132.4 =h 6.4
117.1 =b 3.7

38
39
40
J.1

896-920

15

123.9 =b 3.0

- - 4?

961-980
981-990

11
12

113.2 d= 2.4
109.0 =b 2.7

43

AA

1080-1093

22

115.0 db 2.2

^

1094-1120
1121-1135

14
21

121.7 d= 4.0
120.8 =k 2.9

46

A.7

1270-1310

23

123.4 =fc 3.2

89

90 A. O. WOODFORD

and hence may well represent relatively long periods of nondeposition of clay,
during which the ammonites changed at their usual rates.

At Peterborough there is no sign of erosion, just deposition and nondeposition.
The plasters are uninterrupted sheets, even though just beneath some of them
the clay was reworked by pre-plaster burrowing animals.

Correlations in the Oxford Clay (Callovian) between Peterborough and Kidlington.
The qualitative and quantitative changes of the lower strata groups at Peter-
borough are closely paralleled at Kidlington near Oxford, 70 miles southwest
of Peterborough. The 0-135 cm portion of the Oxford Clay at Peterborough
is the Jason Zone, the lower part of the Middle Callovian. It can be correlated
with 450-500 cm of Oxford Clay that in 1948-51 was temporarily exposed at
Kidlington. At that place, Callomon (1955) collected 165 measurable speci-
mens of the Gulielmites and Guhelmiceras subgenera of Kosmoceras. Brinkmann
had 1086 Kosmoceras specimens (349 measurable) collected from the 0-135 cm
clay at Peterborough, and all but seven were assigned to the same two sub-
genera. The Kosmoceras (Guhelmites and Guhelmiceras) diameters at Peterborough
and Kidlington are similar at corresponding horizons.

The most exact Peterborough-Kidlington correlation is for the faunal break
which occurs at the 51.5-cm plaster in the Peterborough quarries and probably
at the plaster 173 cm above the zero level at Kidlington, although the clay
just below the Kidlington plaster lacks ammonites. Shell sizes for both Guliel-
mites and Guhe Imicerasr are markedly greater above this horizon. Other charac-
teristics change too, especially for Gulielmites^ so that Callomon described the
new species Kosmoceras (Gulielmites) medea (Fig. 1, column 1) to include the
Gulielmites shells below this horizon (boundary of Jason/Medea subzones,
Table 4). There are eight plasters in the Jason Zone at Peterborough and six
at Kidlington. Although the plasters in the Peterborough quarries extend for
miles, the Jason/Medea plaster correlation is the only one that can be made
between Peterborough and Kidlington with any confidence. The plasters mark
times when clay was not being deposited, although ammonite shells were
accumulating, and the area must have been covered by clear sea water. The
failure of most plaster correlations between Peterborough and Kidlington may
mean that most plasters were relatively local; perhaps they were formed on
that side of a large delta which for a time was not receiving muddy water.

In the Oxford-Peterborough belt of Kellaways Rock and Oxford Clay, the
Upper Callovian Athleta and Lamberti zones are not subdivided, but each
Middle or Lower Callovian zone is divided into two or more subzones (Table 4).
The subzones have been recognized in some but not all British Callovian sec-
tions. The Erymnoceras coronatum Zone, i.e., the upper part of the Middle
Callovian, has as its index species a heavy-ribbed ammonite 16-20 in. in diam-
eter that is common in Britain; the zone can also be recognized by the Kos-

CORRELATION BY FOSSILS

91

TABLE 4

CALLOVIAN ZONES AND SUBZONES IN BRITAIN, AFTER CALLOMON (1955)

Zone

Subzone

Upper Callovian Quenstedtoceras lamberti

Peltoceras athleta

Middle Callovian Erymnoceras coronation

Kosmoceras (%ugokosmokeras) grossouvrei
Kosmoceras (ugokosmokeras) obductum

Kosmoceras (Gulielmites)
jason

Kosmoceras (Gulielmites) jason
Kosmoceras (Gulielmites) medea

Lower Callovian Sigaloceras callomense

Sigaloceras planicerclum
Sigaloceras callomense
Proplanidites koenigi

Macrocephalites
(Macrocephalites)
macrocephalus

Macrocephalites (Kamptokephahtes)
kamptus

Macrocephalites (Macrocephalites)
macrocephalus

moceras guides to the two subzones. At Peterborough, the Coronatum Zone
is divided into the Obductum Subzone, from 135.5 to about 854 cm, and the
Grossouvrei Subzone, from about 854 cm to 1093.5 cm. The Kosmoceras guides
to the subzones have been found at Christian Malford in the Oxford Clay belt
35 miles southwest of Oxford, and K. grossouvrei (Table 4), at least, is present
at Weymouth on the south coast (Arkell, 1947, p. 29). The Jason Zone at
Peterborough is divided into the Medea Subzone, 0-51.5 cm, and the Jason
Subzone, 51.5-135.5 cm.

Brinkmann and Callomon emphasized different aspects of the Kosmoceras
evidence. Brinkmann mentioned species and zones, but for him the Guhelmites-
%ugokosmokeras lineage was one unit, in which variations in size, form, and
sculpture were treated statistically. Statistical breaks at plasters were inter-
preted as proofs of lacunas in the record of a continuously evolving population
or sequence of populations. Callomon gave statistics but used them to define
species, zones, and subzones. Callomon's correlations between Peterborough
and Kidlington were correlations of the Medea and Jason subzones, in the
course of which it was shown that the Kosmoceras shells went through similar
changes in adult size at the two localities.

Callovian in Germany, France, and Switzerland. The Callovian is well represented
throughout the Northwest European Province. In the Schwabian and Fran-

92 A. O. WOODFORD

conian Alb of southwestern Germany all six zones are probably present, al-
though the Calloviense Zone (No. 37 in Table I) is poorly represented and
probably incomplete. The Middle and Lower Callovian has the thin "Macro-
cephalenband" at the base, but mostly consists of the clay called Ornatenton,
which is 10 meters thick and richly fossiliferous, with abundant Kosmoceras.
The sequence of Kosmoceras species is confusing and unexplained (Arkell, 1956,
p. 118). The Upper Callovian, fully exposed during the building of the Auto-
bahn in 1937 (R. and E. Model, 1938), is a three-meter clay crowded with
ammonites, many of which are pyritized or phosphatized. Large specimens
of the knobby Peltoceras athleta occur near the base, and specimens of the sharp-
keeled, streamlined Quenstedtoceras lamberti occur at higher levels. An overlying
half-meter of nodular clay contains more or less crushed pyritized or phos-
phatized specimens of both Q. lamberti and the Lower Oxfordian (lowest Upper
Jurassic) guides, forming one of the "condensed" zones that are common in
the continental Jurassic and may be the result of erosion and redeposition of at
least part of the ammonites.

In the Jura tableland, including the region in eastern France around Besan-
on and Belfort north of the Jura Mountains and also extending into Switzer-
land beyond Basel to the Herznach iron mine, all six Callovian zones are
present (Theobald, 1957), although not all are known throughout the area.
The best faunas are found in 10.5-ft deposits of iron ore and associated strata
in the Herznach iron mine (Callomon, 1955, p. 250). Here the Lower Callovian
is present, but the Jason Zone kosmocerids are lacking, and the next fossiliferous
horizons are the well-marked Coronatum and Athleta zones. The successive
faunas are sharply contrasted instead of including closely related kosmocerids
as at Peterborough. At Besangon, a newly exposed Coronatum-Athleta-
Lamberti sequence (Rangheard and Theobald, 1961) includes a Coronatum
Zone with Erymnoceras coronatum and abundant Hecticoceras of Middle Callovian
species, but no Kosmoceras.

In general, on the European continent the distribution of Kosmoceras is very
spotty, but the Callovian zones can be correlated with those in England by
the use of kosmocerids and other guide ammonites. Callovian lacunas are
present on the continent, as in England, and some of these, at several levels,
coincide with slight erosional unconformities and concentrations of battered,
silicified, or phosphatized ammonites.

At lower horizons, in the Bajocian, one or two slight angular unconformities
occur, in England and in Normandy. In Wiltshire, S. S. Buckman (1901)
demonstrated extensive overlap and a discordance of 7 feet per mile.

European Lower and Middle Jurassic zones. The Lower and Middle Jurassic
zones of the Northwest European Province are all of the same general type as
those described for the Callovian. All 41 zones are present in England (column 1
of Table 1). Almost all are present in France, especially on the east side of the

CORRELATION BY FOSSILS 93

Paris Basin and in the Jura tableland (column 2 of Table lj, and in the cuestas
of the Schwabian and Franconian Alb of southwestern Germany (column 3).
Almost all the zones are probably present in the thick strata of the northwest
German basin between the North Sea and the Harz Mountains (column 4)
and in the folded Jura of the Franco-Swiss border (column 5j.

Facies zones. The Lower and Middle Jurassic zones of northwestern Europe
have long been considered time-stratigraphic units. Before we accept this
important inference, we should survey all the other possibilities. Under the
general assumption of organic evolution, there are two other possibilities or
possible complications: (1) local evolution, with lag elsewhere as a result of
poor communications, and (2) migration of relatively long-lived facies faunas.
We shall consider facies faunas first, illustrating them with the present-day
marine facies found off north-south coasts, with special attention to foraminiferal
facies off California because the foraminiferal species living there are also found
fossil in nearby Pleistocene and Pliocene strata.

The present-day shallow-water marine faunal temperature and depth facies
have been thoroughly studied along several north-south trending coasts, in-
cluding the Pacific Coast of North America (for the mollusks, see Burch,
1945-46; Keen, 1958). On the Pacific Coast, breaks between well-defined
molluscan provinces occur at the tip of the Lower California Peninsula, at
Point Conception, California, and elsewhere.

Local and regional differences in the Pleistocene and Pliocene molluscan and
foraminiferal faunas can be established by comparisons with living species.
One kind of facies succession in the rocks is clearly the result of temperature
change. Examples are successions of alternately warm and cold Pleistocene
molluscan faunas at many places along the east and west coasts of North
America. No doubt these faunas moved south during glacial epochs and north
during interglacial periods. Practically all the species are living somewhere
along these coasts today: the cold-water species live in the north and the warm-
water species in the south. But the same species also have a tendency to move
down into deeper colder water somewhat south of their inshore, shallow-water
range. There can be no doubt that these molluscan species, still living today,
are facies fossils in their Pleistocene occurrences, rather than guides to some
particular parts of Pleistocene time, but there may be some uncertainty as to
whether, at any particular latitude, the facies succession is due to climatic
change or to change in the depth of the water in which the Pleistocene animals
lived (Woodring and others, 1946). In either case, the Pleistocene facies suc-
cessions at different localities can be no more than homotaxial and cannot be
used as evidence of synchronism.

In other facies successions, change of depth seems to have been the variable
factor. Examples are found in southern California in two small basins, each
some 25 X 60 miles in horizontal dimensions, where the sediments and faunas

94 A. O. \VOODFORD

have been studied in connection with oil field exploitation. In the Los Angeles
and Ventura basins, the presumably Pliocene strata are 10,000 to 15,000ft
thick and are made up of alternating beds of sandstone and shale, with minor
marginal conglomerate. In both basins, foraminiferal (foramj species are
locally satisfactory guide fossils. The zones occur in the same order in the two
basins, are of comparable thickness, and conform to the structures in the basins,
notably to the folds and faults determined from the sandstone layers that are
the reservoir rocks in the numerous oil fields. But the same foram assemblages
found in the basin strata also occur living on the floor of the nearby Pacific
Ocean, with well-marked depth-of-water ranges. In a way, the order offshore
is the same as in the rocks. The shallowest foram fauna of the oil-well sequence
is also found on the sea floor in shallow water, and the oil-field zones follow in
order down the slope of the sea floor, until the deepest oil-well foram assemblage
is found living on the continental slope of the present ocean at depths greater
than 6000 ft (Natland, 1933, 1957; Bandy, 1953). Almost all the foram species
found in the wells or in Pliocene outcrops are also living offshore today. The
very few extinct species are almost all in the lowest zone in the rocks. All the
zones in the rocks, including the lowest one, must be primarily depth zones,
distinguished by the faunal facies of a particular depth, namely the depth of
water at the time of deposition of the strata involved. Apparently the lowest
strata, those containing several extinct foram species, were deposited in water
more than 6000 ft deep, and all the higher strata were deposited in shallower
and shallower water as the basins filled. The deep-water marine basins appear
to have been steep sided, with very narrow marginal selvages. These selvages
have been involved in deformation and erosion, so that horizons are hard to
trace, but on the east side of the Los Angeles Basin lateral transitions from deep
to shallow facies are probably preserved.

In the two southern California basins, the homotaxis of foram zones does not
necessarily involve synchronism. We do not know whether any particular facies
zone was deposited at the same time in both basins, although the identical
extinct species in the lowest zone of the Pliocene rocks have been considered
sufficient to justify time correlation for this zone.

The weakness of the depth-zone forams as guides to relative age has been
demonstrated by comparisons with the faunas in the Neogene of the Great
Valley of California, 200-300 miles to the north. The strata called Pliocene in
the Valley contain assemblages of shallow-water foraminifera, all assignable to
living species, and are underlain by strata called Miocene. The deeper foramin-
iferal facies zones are missing. It is impossible by the use of foraminifera alone
to develop parallelism between subdivisions of the Pliocene strata of the Great
Valley and the zones set up for the southern California basins.

Facies fossils are harder to recognize in the older rocks. The stratigrapher
feels his way, working back from the present. Facies that merely indicate
depth differences can usually be avoided by looking for forms whose structures

CORRELATION BY FOSSILS 95

(or relationships to living forms of known behavior) suggest the swimming or
floating mode of life. The very fact of wide geographic distribution, essential
for a guide fossil, is suggestive of mobility unlikely in organisms narrowly
restricted in habitat.

In evaluating an extinct kind of guide, such as the ammonites, one must
rely in part on lithofacies and the nonammonitic biofacies. Ammonites are
especially abundant in clays. In the British Callovian they are common in
the Oxford and other clays but are not found in the interbedded sandstone,
the Kellaways Rock. The ammonites of the Oxford Clay are associated with
numerous pelecypods, including oysters. Oysters are commonly shallow-water
forms. In general, ammonites have been found associated with all marine
biofacies except reef corals. They have not been found in brackish, lagoonal,
estuarine, or freshwater associations.

Westermann (1954) recognized at least two alternating ammonite facies
(stephanoid and sonniniid) in the Middle Bajocian calcareous-sandy clays of
the northwest German trough at Alfeld east of Bielefeld, close to the Harz
horst. The stephanoid facies occurs three times, the sonniniid twice, but the
whole sequence can probably be divided satisfactorily between the Stephanoceras
humphriesianum Zone (No. 27 in Table 1) and the Otoites sau&i Zone (No. 26),
since both index fossils are members of the stephanoid group.

On a larger scale, there are two European ammonite facies of provincial
scope, which may be temperature or depth facies. The ammonites of the whole
northwest European platform, including those in the Oxford Clay and also
those in the northwest German trough, may represent a cooler or shallower
facies than that of the Mediterranean Tethys, the home of the smooth, globose
genera called Lytoceras and Phylloceras, which are almost absent from the
platform.

Gayle Scott (1940) attempted to establish five Mesozoic depth zones, which
he lettered from A to E. The deepest zone (E) was the Tethyan zone of Lytoceras
and Phylloceras. The four shallower zones were all illustrated by Texas Creta-
ceous faunas, the shallowest (A) without ammonites. The ammonites of
depth(?) zone D are smoothly rounded shells, found in marls and marly lime-
stones. In the supposed depth zone C, the ammonites are quadrate and highly
sculptured; they occur in marls and clays as well as in chalk and dense lime-
stone. The slender sharp-keeled ammonites of depth (?) zone B (15-20 fathoms?)
are found in sandy limestones and sandy shales.

Stratigraphic paleontologists have a built-in bias against facies interpreta-
tions of their stock in trade, the guide fossils, and Scott's suggestions were not
spontaneously accepted. In this case, caution seems to have been justified, as
ammonites of the supposed depth zone D have been found in a Fort Worth
quarry associated with sea weeds and with genera of foraminifera that today
live attached to plants in shallow water (Claude Albritton, personal communi-
cation).

96 A. O. WOODFORD

Comparison of European ammonite and Calif ornia foram zones. The northwestern
European Jurassic ammonite zones resemble the foram facies zones of the
southern California Pliocene in one way. Both have been found invariably in
the same order. But the Pliocene forams were sedentary bottom animals with
living relatives confined to depth zones. The ammonites were probably active
swimming animals similar to the living Nautilus. Not only did the ammonites
probably get around easily but their empty shells may occasionally have been
spread rather widely by currents, just as a few nautilid shells have been carried
hundreds of miles to Japan, far northeast of the haunts of the living animals.
Finally, the unfailing ammonite homotaxis in the large European Jurassic area
contrasts with the failure of Pliocene foram homotaxis between the southern
California basins and the Great Valley of central California.

Possible faunal lag in the European Jurassic. In the northwest European Lower
and Middle Jurassic there is little or no evidence for local lag in faunal change
at a specific horizon, or series of horizons, as a result of temporary barriers or
the time required for migration. The practically complete homotaxis through-
out the province, zone by zone, combined with the Peterborough evidence for
an evolutionary sequence, is explainable by the easy migration of rapidly
evolving animals and probably in no other way. If this explanation is the
correct one, the homotaxis implies time correlations.

Before accepting this conclusion of synchronism for each zone throughout
the province, we should try to discover the implication for travel times ex-
pressed in years. Since the time of deposition for the whole Jurassic System
was probably of the order of 45 million years (compare Holmes, 1960; Kulp,
1961) and the system is divided into upwards of 60 zones (41 for the Middle
and Lower Jurassic, 20 or more for the Upper), the average zone accumulated
in about 750,000 years. Modern marine clams and snails have spread through
scores or even hundreds of miles of shallow water in a few decades (Elton,
1958). If the ammonites got around as quickly as the clams and snails do now,
which seems likely, the spread of a new European zonal fauna must have been
geologically instantaneous. The conclusion follows that the Lower and Middle
Jurassic zones really are temporal units throughout the Northwest European
Province.

Extent and meaning of lacunas in the northwest European Lower and Middle Jurassic.
If the northwest European Lower and Middle Jurassic zones are time-strati-
graphic units within the province, what are the extent and meaning of the
frequently sharp zonal boundaries? Brinkmann and Callomon showed that
in Britain the Coronatum-Jason boundary of the Middle Callovian marks a
lacuna in the paleontologies! record. Apparently this boundary also coincides
with a paleontological lacuna at many places in the continental portion of the

CORRELATION BY FOSSILS

97

TABLE 5

LOWER JURASSIC CORRELATIONS: NORTHWEST EUROPE-MEXICO (ERBEN, 1956)

Stage

Northwest European zone
(Table 1)

East-central Mexico ammonite
"faunizones"

Pliensbachian 10. Uptonia jamesoni

Uptonia sp.

Sinemurian 9. Echioceras raricostatwn

8. Oxynoticeras oxynotwn
7. Asteroceras obtusum

6. Caenisites turneri

5. Arnioceras semicostatum

4. Arietites bucklandi

(Coroniceras subzones)

(Microderoceras bispinatum altespinatum
[Ecfuoceras burckhardti
(Pleurechioceras? james-danae
[Pleurechioceras subdeciduum
? ?

Vermiceras bavaricum mexicanum*

Oxynoticeras sp.f

? ?

(Euagassiceras subsauzeanum
(Arnioceras geometricoides

Coroniceras pseudolyra

* Doubtful correlation.

t Perhaps equal to European zone 8.

province, with at least local disconformity and the reworking of ammonites.
However that may be, the kosmocerids must have lived right through the time
of restricted deposition, in some part or parts of the region. Northwestern
Europe was their homeland, and they were never abundant elsewhere. At
other zone boundaries exotic guide fossils appear, perhaps as immigrants from
Tethys (compare Spath, 1933, p. 427) and disappear, perhaps becoming
extinct without issue.

Worldwide significance of northwest European Jurassic zones. No northwest
European Lower or Middle Jurassic zone is worldwide in extent. Neverthe-
less, many of the zones have been recognized at one or more places outside the
province (compare Table 1), through the presence of the index species of the
zone, other guide species, or closely related species. Take, for example, the
Lower Jurassic correlations between Europe and eastern Mexico (Table 5).
Note that zones 4, 5, 9, and 10 of the European sequence are confidently
correlated with local Mexican zones ("faunizones") although no index (zone
name) species, and almost no ammonite species, is common to the two regions.
Some Mexican species, however, are very similar to guides for the European
zones; examples are Echioceras densicosta of Mexico, with about 25 costae on
the last whorl, compared to 18 for the index species of the Echioceras raricostatum

98 A. O. \VOODFORD

Zone, and Coromceras pseudolyra of Mexico, similar to Coromceras lyia of the
Anetites bucklandi Zone. There are difficulties at the horizons of zones 6, 7,
and 8 (Table 1). Nothing like zone 6 has been recognized in Mexico. The
author of Table 5 (Erben, 1956) seriously considered correlating zone 8 (Oxy-
noticeras oxynotum) of northwestern Europe with the Oxynoticeras sp. Zone of
Mexico, but finally chose the zone 7 (Asteroceras-Vermiceras] correlation, shown
in Table 5. This drops Oxjnoticeras two notches. The correlation of Table 5
or any other possible set of matches leaves two zones in each region unmatched
in the other.

Actually, the matching of zones between eastern Mexico and northwestern
Europe is exceptionally close. In general, intercontinental correlation, zone
by zone, is not practical. For most horizons, larger and more generalized units
substages must be used.

Lower and Middle Jurassic Stages and Substages

Standard stages and substages of northwestern Europe. The zones of Table 1 are
grouped into stages. Although stages were established before zones, a stage
is most precisely defined in terms of the zones that make it up in its type area,
which for the Lower and Middle Jurassic is northwestern Europe. Since the
stages are based on zones, they too are defined on a paleontological basis.
Stages have boundaries that may cross formation boundaries obliquely, even
in the type areas; stages may be represented by different lithologic facies in
different areas; stages can be identified in isolated distant places all because
they are based on assemblages of guide fossils. In 1956, just before his death,
the British stratigrapher W. J. Arkell (1956, p. 9) wrote of the Jurassic stages 2 :
"As units of the single world scale of classification, stages must be based on
zones . . . They are essentially groupings of zones, but they transcend zones
both vertically and horizontally." Vertically, a sequence of zones makes one
stage. Horizontally, the characteristic zone assemblages of fossils may disap-
pear, but the stage may still be recognizable and still divisible. For distant
correlations, therefore, both stages and substages are needed. A substage is a
major subdivision of a stage, defined in terms of the zones that make it up in

2 The Middle Jurassic stages were based fairly closely on William Smith's formations.
The Callovian Stage was named from Smith's Kelloways Stone, by Latinization; it also
includes most of the overlying Oxford Clay. The Bathonian Stage includes Smith's
Great Oolyte, or Bath Freestone, named after Bath, England, and also the Forest
Marble and Clay. The Bajocian is Smith's Under Oolyte, but the type locality is across
the Channel at Bayeux, Normandy. The four Lower Jurassic stages were all carved
out of Smith's Lias clay, shale, and limestone, but the type localities are in France
and Germany. Unlike the zones, the stages have remained practically unchanged since
1864.

CORRELATION BY FOSSILS 99

the type area. Each Middle Jurassic stage is divided into upper, middle, and
lower substages. The three higher Lias stages Toarcian, Pliensbachian, and
Sinemurian have upper and lower substages; the Hettangian is not divided.

Recognition of Callovian and other stages outside Europe. Sample Middle and Lower
Jurassic sections in Asia and America are shown in Table 1, columns 6-13.
The Callovian is the only stage represented in all 8 columns. The three lowest
stages are found together in two and probably in three of the 8 columns, and
in these two or three they are rather fully represented. The Caucasus Moun-
tains (column 6) and the central and southern Andes (column 13) have excep-
tionally complete sections. Note particularly the number of European zones
that have been identified in the Andes, two-thirds of the way around the globe.

The Callovian Stage, at the top of the Middle Jurassic, is both widespread
and varied. Several Callovian zones are recognizable outside western Europe,
notably in the Caucasus and in Cutch, India (Table 1). Cutch, at the north-
eastern edge of the Arabian Sea, has perhaps the most fossiliferous and paleon-
tologically complex sequence of the lower part of the Callovian in the world.
Close correlation with Europe is difficult because the Cutch faunas are com-
posed largely of Oriental elements. Kosmoceras is unknown there.

In western North America the Lower Callovian substage has been reported
in several areas. In the Wyoming-Montana area five regional zones have been
distinguished (Imlay, 1953). In this same region a zone with Quenstedtoceasr
may be approximately equivalent to the Quenstedtoceras lamberti Zone of Europe,
the highest Callovian zone in the type section (Table 4).

In the southern hemisphere some of the most characteristic European Cal-
lovian genera do not exist, but the Callovian Stage can be recognized "by the
general grade of evolution of the ammonite fauna as a whole and by a chain of
overlapping correlations carried link by link round the world" (Arkell, 1956,
P- 12).

Time correlation from stage and substage homotaxis* In the statement just quoted,
Arkell assumed (1) that the presence of the same guide ammonites at two
localities provides sufficient evidence for synchronism, and (2) that local
"overlaps" of provincial ammonite taxa are almost equally good evidence for
synchronism. Similar assumptions are made by most, but not all, stratigraphic
paleontologists. We must now consider whether the assumptions are justified.

Possibilities other than synchronism are hard to formulate in a general and
exhaustive way, especially because the data vary from horizon to horizon.
With respect to the ammonite guides of the Lower and Middle Jurassic stages
and substages, we shall assume that local facies variations have been accounted
for at the zone level. Facies of provincial scope do not provide homotaxial
problems. The possibilities for significant error at the substage level appear

100 A. O. WOODFORD

to be two: (1) difference in the time range of the substage guides in two or more
districts or provinces, as by survival in a distant province after extinction in
the homeland, or vice versa, and (2) misinterpretation of differences in details
or rate of evolution in two provinces separated by a barrier. The questions at
issue are the probability and magnitude of these sources of error. Every case
must be treated on its own merits, although the cumulative effect of evidence
for synchronism at horizon after horizon would be to establish a presumption
favorable to synchronism for other cases.

We shall take the Toarcian Stage of the Lower Jurassic as an example.
Typical Toarcian faunas can be followed around the world in the northern
hemisphere, from western Europe to the Donetz Basin of southern Russia, the
Caucasus, Persia, Baluchistan, Indonesia, Indochina, Korea, Japan, Alaska,
a Canadian Arctic island, eastern Greenland, and back to northwestern Europe
(Arkell, 1956). Everywhere European guide genera are found and in many
places European guide species. Grammoceras thouarsense is a guide fossil for a
zone in the midst of the stage in the type section at Thouars, France. The same
species is found in the same homotaxial position in the Toarcian of the Cau-
casus. Other west European guides also occur in the Caucasus, in about the
standard order, although the thick section has structural complexities and
definitive studies have not yet been made. In Japan the Toarcian zones are
almost the same as in western Europe, with typical European genera but
peculiar Japanese zone species. In eastern Greenland the Upper Toarcian
and a Lower Toarcian zone (No. 17) can be distinguished, both with guide
species similar to those in Europe (Callomon, 1961). The Toarcian localities
range in latitude from the equator (Indonesia) nearly to the North Pole (the
Arctic island). The facies range from platform to thick geosynclinal with
interbedded volcanics (Caucasus). In Japan some 1500 feet of Toarcian sand-
stone and shale are known, and in the Caucasus 15,000 to 20,000 feet of sand-
stone and shale, interbedded with freshwater strata containing coal.

Let us imagine that the two successive Toarcian faunas that can be generally
recognized (Lower and Upper) originated in northwestern Europe and mi-
grated eastward, and only eastward, around the world, evolving slowly as
they migrated, and that after circumnavigation the lower fauna had evolved
into the upper. By the time the migrants reached Japan new species might
well have developed. To this extent the hypothesis fits the evidence. But
evidence for eastward migration ends with Japan. The Japanese species did
not get to Alaska or Greenland. The somewhat scanty evidence suggests a
different hypothesis, namely, migration both east and west from an evolu-
tionary center that may have included both Tethys and the European plat-
form, with development of local species in distant, more or less isolated areas,
such as Japan, but with communication sufficiently complete and rapid to
permit the earlier Toarcian fauna to be everywhere overwhelmed by the

CORRELATION BY FOSSILS 101

second soon after the second had become well characterized anywhere. This
hypothesis explains the similar succession of Toarcian genera all around the
world, in Japan as well as in Europe. That is to sa\, the evidence favors inter-
pretation of the Toarcian homotaxis in Europe, Japan, etc., as evidence for
the time correlation of the highly varied enclosing strata, substage by substage,
or even in some places zone by zone.

The Toarcian is a specially instructive stage, for one must work around the
world with some care in order to find at all localities the same typical European
Lower Toarcian genera. A more southern course from Europe to the Indian
Ocean yields a somewhat different set of results for the Lower Toarcian. This
route goes from the northwestern European Platform to Portugal and then
jumps to Arabia, East Africa, and Madagascar, where, on the south side of
Tethys, the peculiar Bouleiceras fauna, unlike anything in northwestern Europe,
characterizes a province called Ethiopian by Arkell (1956, p. 614). In Portugal,
Bouleiceras itself occurs in the same beds with northwestern European guides;
this is the "overlap." The Ethiopian Lower Toarcian fauna has Oriental
affinities that perhaps make possible correlations with Baluchistan and eastern
Asia, but this correlation is not so well established as those on the northern
route. One cannot assert positively that the Lower Toarcian of Arabia is the
time equivalent of the Lower Toarcian of northwestern Europe, but one can
say that this time correlation is more likely than any other.

Upper Jurassic Provincialism

Ammonite provincialism increases in the Upper Jurassic. Correlations in
the lowest Upper Jurassic stage, the Oxfordian, are achieved in the same way
as for the Callovian (Arkell, 1956, p. 12). Worldwide correlations are also
possible for the lower part of the next higher stage, the Kimeridgian. In the
remainder of the Upper Jurassic, however, three European provinces are only
too well defined: northwestern European, Tethyan, and Russian- Arctic, or
Boreal. The uppermost Jurassic faunas which are most satisfactory for world-
wide homotaxis and time correlation are those of the Tithonian Stage of
western Tethys (Gignoux, 1955, p. 354), with type localities in or near the
French Alps. For these uppermost Jurassic levels, northwestern Europe cannot
furnish a standard for the world.

The Russian-Arctic faunas may reflect a cold-water facies, although some of
the genera are known from widely separated parts of the great Pacific region.
In some parts of this region ammonites are rare or absent and uppermost
Jurassic-lowest Cretaceous correlations are made by using species of the oyster-
like genus Aucella (also called Buchia). The shells of several species of Aucella
are abundant in European Russia and California and are known in Mexico
and New Zealand.

102 A. O. WOODFORD

The Upper Jurassic faunas of the Northwest European Province may repre-
sent the alternation of cold-water and warm-water conditions. Northern
ammonite genera appear in the Mediterranean at the base of the Oxfordian,
and Upper Oxfordian coral reefs are present as far north as central England.

Once the provinces become sharply defined, correlation between them be-
comes difficult but not impossible. Worldwide Upper Jurassic correlations may-
yet be achieved, but the present situation still makes impressive, by contrast,
the well-established correlations of the Lower Jurassic and part of the Middle
Jurassic.

Top and Bottom of Jurassic

Boundaries in type section. In southern England, the marine Jurassic forma-
tions, with guide ammonites, form a natural unit, bounded above and below
by beds transitional to nonmarine formations. The uppermost English Jurassic
unit, the brackish and freshwater Purbeckian, without ammonites, is overlain
by the continental Wealden beds of the Weald anticline, south of London,
which are called Cretaceous. The basal transition beds, called the Rhaetic
(or Rhaetian), are made up largely of shale and limestone containing pectens
and other shallow marine mollusks, but also include the Rhaetic bone bed
with its abundant remains of fish, amphibians, and reptiles. The Rhaetic has
its type area in southern Germany and northern Tyrol; there, in the north-
eastern Alps, it contains a few ammonites of Triassic aspect that justify its
incorporation in that system. In both England and Germany the Rhaetic
is overlain conformably by the Psiloceras planorbis Zone, zone 1 of the Jurassic.

Extent of Planorbis %pne. The lowest Jurassic is recognizable as the Planorbis
Zone throughout the northwest European platform area, and in England and
France, it is divisible into two subzones. The Planorbis Zone has also been
recognized in Sicily, western British Columbia (West Canada in Table 1), and
Peru. Beds containing species of the guide genus Psiloceras, and hence probably
at the approximate Planorbis Zone horizon, are also known in western Nevada,
New Zealand, New Caledonia, and (from boulders) on Timor. In central
Europe and Nevada, Planorbis strata overlie beds with characteristic upper-
most Triassic ammonites.

Precision of the base of the Jurassic. The base of the Planorbis Zone, lowest in
the Jurassic, seems to be a fair example of a fossil-determined Mesozoic horizon
that is recognizable in several continents. We shall make estimates of the prob-
able precision of this horizon, in years and in percent of its age. We take the
probable average span of a Jurassic zone as 750,000 years, as developed on a
previous page. Judging by the ranges found for the zone guides on the European

CORRELATION BY FOSSILS 103

platform, the Jurassic guide species did not last longer than the time repre-
sented by a European zone. If the Planorbis Zone took twice the average time
for its deposition, that is, about 1.5 million years, the age of the base of the
Jurassic, where the Planorbis Zone is present, probably does not vary by more
than this length of time. Since 180 million years is a reasonable guess for the
radiometric age of the base of the Jurassic (Holmes, 1960; Kulp, 1961), the
base of the Planorbis Zone, wherever it is present, can be correlated, even from
continent to continent, with a probable precision of one percent or better,
even though this estimate is obviously not based on standard statistical pro-
cedures.

Triassic and Cretaceous Homotaxis and Correlations

The Triassic and Cretaceous systems have been zoned, and the zones have
been grouped into stages in about the same manner as was done for the Jurassic.
Ammonites are the principal marine guide fossils. Homotaxis, supplemented
by overlapping correlations, is worldwide for large parts of each system. The
marine Cretaceous is considerably more widespread than the Jurassic, the
marine Triassic considerably less. The type area for the Triassic is in the
eastern Alps, where 6 stages have been established (Brinkmann, 1954, p. 175).
The type sections for the 12 Cretaceous stages now commonly recognized are
mostly in France, although some are just outside that country (Gignoux, 1955,
pp. 392, 398; see also "Danian," below, under Cenozoic). In many regions
the Upper Cretaceous contains few ammonites. Echinoids, forams, pelecypods,
and gastropods must then take the ammonites' place, especially for local cor-
relations. Study of the ammonite (and planktonic foraminiferal) faunas indi-
cates that correlations to the substage level are generally justified, as in the
Jurassic.

Conclusions Concerning Mesozoic Stages

As a summary of our consideration of Mesozoic stages and their ammonite
faunas, the following statements may be made.

1. Mesozoic stages are groupings of the rock sheets called zones, based
on ammonite assemblages.

2. Although zones are more or less local, Mesozoic stages and even
substages are mostly worldwide, made so by step-by-step correlations of
geographically changing ammonite assemblages.

3. Mesozoic stages and substages are time-stratigraphic units: each
stage or substage is composed of strata that accumulated at about the same
time in all the places where the stage or substage has been properly identi-
fied.

104 A. O. WOODFORD

This statement is vitally important, but it contains two ambiguous phrases,
"about the same time" and "properly identified," which need to be reempha-
sized. The second is easier to handle. A proper identification should involve
the use of ammonites, whose stratigraphic range should be connected with the
ranges of the guide ammonites in the pertinent European stages by methods
indicated on previous pages. "About the same time" expresses confidence that
there is little time overlap of the stages (or substages). For example, we think
that practically all the marine Sinemurian of all regions was deposited before
the accumulation of any ammonite-bearing strata anywhere that have been
assigned to the Pliensbachian. The degree of our confidence and the reasons
for it have been stated in such paragraphs as those on the European-Mexican
Lower Jurassic homo taxis (compare Table 5).

Paleozoic Guide Fossils and Correlations

For the Paleozoic, no single group of animals has provided an adequate set
of guide fossils. Goniatitic ammonoids are useful in the Upper Paleozoic, but
they are less numerous and less widely distributed than their Mesozoic relatives.
Goniatite zones have about the same significance as Mesozoic stages. Zones
almost equally valuable may be established by using other organisms. Trilo-
bites are outstanding for the Cambrian, graptolites for the geosynclinal facies
of the Ordovician and Silurian, and fusuline forams in the Pennsylvanian and
Permian. In the highly fossiliferous and widespread limestones and shales of
the broad continental platforms, the most useful guide fossils, especially for
relatively short-distance correlations, are brachiopods.

Paleozoic subsystems and stages. Each Paleozoic system below the Carboniferous
is divided into thirds, and these subsystems have worldwide correlative signi-
ficance. Below the Devonian, stages or other worldwide units smaller than
subsystems seem to be in uncertain status, at least temporarily. Recently, Bell
(1960) and Berry (1961) have emphasized the difficulties of interprovincial
stage correlations in the Cambrian and Ordovician. In the marine Devonian,
useful stages have been established, with type areas in Belgium or the Rhine-
land. Gignoux (1955, p. 120) gave the following stages, from the base up:
Gedinnian and Coblencian in the Lower Devonian, Eifelian and Givetian in
the Middle Devonian, Frasnian and Famennian in the Upper Devonian.
Brinkmann (1954) gave a somewhat different list, with 9 stages. The Car-
boniferous is divided into Lower (also called Dinantian or Mississippian,
although these two terms are not synonymous; compare Brinkmann, 1954,
opposite p. 106) and Upper (Pennsylvanian or Coal Measures). The Upper
Carboniferous is divided into three stages designated, in ascending order,

CORRELATION BY FOSSILS 105

Namurian, Westphalian, and Stephanian, with the most satisfactory type sec-
tions (probably even for the Stephanian) in Belgium, west central Germany,
and central Russia because of the presence of marine bands there. Each stage
has characteristic goniatites and the upper two have guide fusulines (Brink-
mann, 1954, opposite p. 106). For the Permian, Brinkmann (1954, opposite
p. 132) gave a composite set of five Old World stages, with guide ammonoids
and fusulines From the base up, with the West Texas equivalent in paren-
theses, Brinkmann's stages are Sakmara (Wolfcamp), Artinsk (Leonard), Sosio
(Word), Basleo (Capitan), and Chideru (Ochoa). A composite set of stages is
unsatisfactory, but the type Permian in European Russia is brackish to non-
marine in its upper part.

In North America, the whole Paleozoic has been provided with more or less
independent local stages and other units (see C. O. Dunbar and others, 1942,
1944, 1948, 1954, 1960), with American type sections or localities.

Cenozoic Correlations

The first problem connected with the Cenozoic is its definition, and the
definition of its marine subdivisions, in the European type region. Preva-
lent current practice seems to be to make the Danian of Denmark a part of the
Cenozoic and put the top of the Mesozoic at the top of the Maestrichtian Stage
in Holland and Denmark (Loeblich and Tappan, 1957, p. 1113; Bramlette
and Sullivan, 1961, p. 136). Few if any worldwide marine Cenozoic stages are
yet valid, although prospects are bright for success in the Paleogene, through
the use of planktonic forams and nannoplankton (Loeblich and Tappan, 1957;
Bolli, 1959; Bramlette and Sullivan, 1961). The Cenozoic series Paleocene,
Eocene, Oligocene, Miocene, Pliocene, and Pleistocene are now well estab-
lished, at least as names, although there is considerable variation in the number,
terminology, and distribution of the stages that make up the series in the
European type section (compare Brinkmann, 1954, p. 254, with Gignoux,
1955, pp. 472 and 557). The base of the Miocene is put variously at the base
of the Aquitanian, at the Burdigalian-Aquitanian boundary, and at the top
of the Burdigalian. For Brinkmann it is sub-Aquitanian, for Gignoux sub-
Burdigalian. Drooger (1954, 1956), within two years, dropped the base of the
European Miocene from the top of the Burdigalian to the base of the Aquitanian.
Overseas correlations are even less definite. Recently Eames and others (1962)
suggested that the Vicksburg and certain other North American formations,
commonly considered pre-Aquitanian Oligocene, should be raised to positions
in the Miocene. In general, Cenozoic intercontinental marine correlations are
exceptionally difficult. The difficulties are magnified if correlatipns involving
nonmarine mammals are taken into account.

106 A. O. WOODFORD

Divisions of Geologic Time

Systems and stages are divisions of the rocks. If we agree that the procedures
by which sedimentary rocks all over the world are assigned to particular systems
and subsystems, or even to stages and substages, are, at least in favorable cases,
correlations of contemporaneous strata, we can make the transition from the
stratigraphic division of the rocks to the division of geologic time. We can
then recognize geologic periods and the primary subdivisions of the periods:
Early Cambrian time, the time of deposition of the Lower Cambrian rocks;
Middle Cambrian time, the time of deposition of the Middle Cambrian rocks;
Late Cambrian time, the time of deposition of the Upper Cambrian rocks; etc.
Some finer time subdivisions can also be made, especially in the Mesozoic.
Callovian time was the time of deposition of the Callovian rocks; early Callo-
vian time was the time of deposition of the Lower Callovian rocks.

Most periods seem to have been between 40 and 70 million years long
(compare Holmes, 1960; Kulp, 1961). Subperiods were therefore very roughly
15-25 million years each. The Cambrian subperiods were probably longer;
a Chinese Lower Cambrian fauna might have an age uncertainty of 35 million
years from vagueness of correlation alone. The Mesozoic stages are more pre-
cise. There are about 11 Jurassic stages and about 12 Cretaceous stages. If
the Jurassic and Cretaceous periods were respectively 45 and 65 million years
long (compare Holmes, 1960; Kulp, 1961), each Jurassic or Cretaceous stage
represents, on the average, 4 or 5 million years. The earliest Jurassic stage,
the Hettangian, which includes only three zones, may stand for a shorter time.
Most Mesozoic substages are thirds or halves of stages and so may represent
1.5 to 2.5 million years each.

Lacunas and the Geologic Column

Recognition and naming of lacunas. Each lacuna in the rock succession repre-
sents a lapse of time. The time represented by the lacuna at 135.5 cm Peter-
borough, between the Jason and Coronatum zones of the Middle Callovian,
is part of Middle Callovian time. If, however, one wished to subdivide Middle
Callovian time, should the separation be into two parts Jason time and
Coronatum time or into three parts Jason time, lacuna time, and Corona-
tum time? If one is considering a global time scale, this particular problem is
unreal, because the European zonal stratigraphy cannot be extended through-
out the world. But similar questions have been raised with regard to pairs of
periods. Should intervals between periods be recognized? Moreover, if a
large lacuna existed between the type Cambrian and Ordovician of Wales or
in the more easily handled Cambro-Ordovicfan section in Scandinavia, should
an American stratigrapher with a more complete section insert a new unit
between Cambrian and Ordovician? This was in part what Ulrich tried to

CORRELATION BY FOSSILS 107

do with his Ozarkian System (Ulrich, 1911, p. 608). A consensus of opinion
now seems to have developed with regard to lacuna fillings. Xew systems are
not to be established. Stratigraphic units with lacuna-filling faunas are to be
assimilated into the systems and subsystems of the long-established standard
column. Just how to do this is a problem of adjustment or negotiation in each
case. To take one example, the marine strata that seem to belong between the
highest Russian marine Permian and the lowest Alpine marine Triassic are
assigned to the Permian, not to the Triassic (Brinkmann, 1954, opposite p. 132).
The question of "intervals" remains. Are there still any unfilled large
lacunas? Unconformities marking lacunas on the continental platforms are
no longer pertinent, now that basins near the margins of the platforms have
been found to contain more continuous sections. Studies of surface and sub-
surface sections in the basin areas have filled in most of the more obvious
lacunas of the fossiliferous marine section. Transitions have been found from
Precambrian to Cambrian (southeast California Nelson, 1962), from Permian
to Triassic (Newell, 1962), and from Cretaceous to Tertiary (central California
Payne, 1951; Schoellhamer and Kinney, 1953). It now seems probable that
somewhere, on some continent or island, or beneath some small or moderate-
sized sea, sedimentary strata exist that formed during the time represented by
every paleontologically measurable lacuna in the European record.

Found changes and lacunas. It should be noted that not every faunal change
is proof of a paleontological lacuna. Even where the ecological facies are
similar or identical and whole groups of animals disappear at one horizon, as
do the ammonites and plesiosaurs at the Cretaceous-Tertiary boundary in
California, the Stratigraphic and paleontological lacuna may be small. On the
west side of the San Joaquin Valley (Schoellhamer and Kinney, 1953), the
shales and sandstone of the Moreno formation are overlain with slight uncon-
formity by the sandstones, mudstones, and claystones of the Lodo formation,
which is glauconitic at the base, but the Cretaceous-Tertiary contact, deter-
mined faunally, is in the Moreno formation more than 200 ft below the uncon-
formity. The provincial Paleocene guide fossils Flabellum remondianum Gabb, a
coral, and Brachysphingus sinuatus Gabb, a gastropod, were collected from an
80-ft sandstone lens in the Moreno, and the top of the lens was about 200 ft
below the Lodo contact. The Moreno shales, 600 ft or more below the sand-
stone lens, contain Maestrichtian ammonites and several kinds of large Meso-
zoic reptiles: plesiosaurs, a mosasaur, and the dinosaur Trachydon. No physical
Stratigraphic break was found between the lowest Paleocene guide fossils and
the highest Cretaceous guides. Regionally, the abundant marine forams of the
highest Cretaceous strata are so similar to those of the lowest 'Paleocene that
the assignment of some faunas to the Cretaceous or the Paleocene is somewhat
uncertain (compare Goudkoff, 1945, p. 1004).

108 A. O. WOODFORD

Summary Discussion

Biostratigraphic generalizations. Stratigraphic correlations through the use of
guide fossils are based on repeated observations over a period of 100-150 years,
with results so uniform that they may be summarized as empirical generaliza-
tions: (1) The fossil species and genera in a single column vary from horizon
to horizon. (2) Homotaxis occurs between columns. (3) Within a province,
guide species useful in time correlation can be established for horizon after
horizon, using the methods described in this paper in connection with the
northwestern European Lower and Middle Jurassic zones. (4) More distant
correlations, between provinces, can be made through step-by-step correlations,
using gradually varying assemblages. Time correlations based on homotaxial
evolutionary sequences will become more and more easily distinguishable from
the misleading correlations of facies zones as more is learned about the evolu-
tion and geologic history of the families involved. Most of the older studies on
ammonite groups were weakened by the use of the now discredited biogenetic
law: ontogeny recapitulates phylogeny. The kind of study needed is illustrated
by Brinkmann's work on the Kosmoceratidae, although that work now deserves
the compliment of restudy and revision. The extension of such work will be
difficult because so few families have their Peterboroughs.

Correlations based on the appearance or disappearance of major groups,
such as trilobites, ammonites, or mammals, are inherently unsafe (Huxley,
1870). Consider, for example, the absurdity of a correlation based on the first
appearance of eutherian mammals in North America and Australia. This rule
does not prevent an ammonite species or genus from being a sound guide to
the uppermost Cretaceous.

In favorable circumstances, such as those that prevail for the northwest
European Lower and Middle Jurassic, time correlations within a province may
be very precise. For significant comparisons with radiometric ages, however,
more nearly worldwide interprovincial correlations must be considered. The
precision of interprovincial correlations varies from one part of the geologic
column to another. In the Jurassic and Cretaceous a time correlation based on
numerous well-preserved specimens of ammonite guide species can hardly be
in error by more than one substage, perhaps =fc2 million years. The base of
the fossiliferous Lower Cambrian, on the other hand, may vary in age by tens
of millions of years.

Stratigraphy and time. Geologic stages and systems are second-order units and
geologic periods are, in a way, third-order units. Stratigraphy began with the
local columns of one region, from which the generalized column of stages and
systems has been chiefly derived. The standard time scale is derived directly
from the standard column and from no other source, except for Late Pleistocene
details. The fossils of the units in the standard column and of other units in

CORRELATION BY FOSSILS 109

other columns are still our principal guides in stratigraphic correlation, al-
though we cordially welcome the statistical calibration of the standard column,
in years, from radiometric data.

REFERENCES CITED

ARKELL, W. J., 1947, The geology of the country around Weymouth, Swanage, Corfe,

and Lulworth: Great Britain, Geol. Survey Mem., 386 pp.

, 1956, Jurassic geology of the world: Edinburgh, Oliver and Boyd, 806 pp.

BANDY, O. L., 1953, Ecology and paleoecology of some California foraminifcra: J. Pale-

ont., vol. 27, pp. 161-182, 200-203.

BELL, W. C., 1960, Cambrian, in McGraw-Hill Encyc. Sci. Techn.: New York, McGraw-
Hill, pp. 424-427.
BENTZ, A., 1928, Uber Strenoceraten und Garantianen insbesondere aus dem mittleren

Dogger von Bielefeld: Jahrb. Preuss. geol. Landesanst., pp. 138-206.
BERRY, W. B. N., 1961, Chorology, chronology, and correlation: Geol. Soc. Am., pro-
gram Ann. Mtg., pp. 10A-11A.
BOLLI, H. M., 1959, Planktonic foraminifera as index fossils in Trinidad, West Indies

and their value for worldwide stratigraphic correlation: Eclogae Geol. Helvetiae,

vol. 52, pp. 627-637.
BRAMLETTE, M. N., and SULLIVAN, F. R., 1961, Coccolithophorids and related nanno-

plankton of the early Tertiary in California: Micropaleont., vol. 7, pp. 129-188.
BRINKMANN, R., 1929, Statistisch-biostratigraphische Untersuchungen an mitteljuras-

sischen Ammoniten uber ArtbegrifF und Stammesentwicklung: Abh. Ges. Wiss.

Gottingen, N. F., vol. 13, no. 3 3 249 pp.

, 1954, Abriss der Geologic: Stuttgart, Enke, 2 vols.

BUCKMAN, S. S., 1901, Bajocian and contiguous deposits in the North Cotteswolds: The

Main Hill-mass: Geol. Soc. London, Quart. J., vol. 57, pp. 126-155.
BURCH, J. Q., ed., 1945-1946, Distributional list of the West American marine mollusks

from San Diego, California, to the Polar Sea: Conch. Club Sou. Calif. Minutes

nos. 45-63, separately paged.
CALLOMON, J. H., 1955, The ammonite succession in the lower Oxford Clay and Kella-

ways Beds at Kidlington, Oxfordshire, and the zones of the Callovian Stage: Roy.

Soc. London, Phil. Tr., s. B., vol. 239, pp. 215-264.
, 1961, The Jurassic System in East Greenland, in Geology of the Arctic, G. O.

Raasch, ed.: Toronto, Univ. Toronto Press, vol. 1, pp. 258-268.
DEAN, W. T., DONOVAN, D. T., and HOWARTH, M. K., 1961, The Liassic ammonite

zones and subzones of the North-West European Province: British Mus. (Nat. Hist.),

B., Geol., vol. 4, no. 10, pp. 437-505.
DROOGER, C. W., 1954, The Oligocene-Miocene boundary on both sides of the Atlantic:

Geol. Mag., vol. 91, pp. 514-518.
, 1956, Transatlantic correlation of the Oligo-Miocene by means of foraminifera:

Micropaleont., vol. 2, pp. 183-192.

110 A. O. \VOODFORD

DUNBAR, C. O., chm., 1942-1960, Correlation charts prepared by the Committee on

Stratigraphy of the National Research Council: Geol. Soc. Am., B., vol. 53, pp.

429-433; Cambrian, vol. 55 (1944), pp. 993-1003; Ordovician, vol. 65 (1954), pp.

247-298; Silurian, vol. 53 (1942), pp. 533-538; Devonian, vol. 53 (1942), pp. 1729-

1793; Mississippian, vol. 59 (1948), pp. 91-196; Pennsylvanian, vol. 55 (1944), pp.

657-706; Permian, vol. 71 (1960), pp. 1763-1805.
EAMES, F. E., BANNER, F. T., BLOW, W. H., CLARKE, W. J., and Cox, L. R., 1962,

Fundamentals of mid-Tertiary stratigraphical correlation: Cambridge, Eng., Cam-
bridge Univ. Press, 163 pp.
ELTON, C. S., 1958, The ecology of invasions by animals and plants: London, Methuen,

181 pp.
ERBEN, H. K., 1956, El Jurasico inferior de Mexico y sus amonitas: XX Int. Geol.

Congress, Mexico, 393 pp.
GIGNOUX, M., (1950) 1955, Stratigraphic geology (trans, by G. G. Woodford): San

Francisco, Freeman, 682 pp.
GOUDKOFF, P. P., 1945, Stratigraphic relations of Upper Cretaceous in Great Valley,

California: Am. Assoc. Petroleum Geol., B., vol. 29, pp. 956-1007.
HOLMES, A., 1960, A revised geological time-scale: Edinburgh Geol. Soc. Tr., vol. 17,

pp. 183-216.
HUXLEY, T. H., 1862, The anniversary address: Geol. Soc. London, Quart. J., vol. 18,

pp. xl-liv.
, 1870, Anniversary address: Geol. Soc. London, Quart. J., vol. 26, pp. xxix-

Ixiv.
IHLAY, R. W., 1953, Callovian (Jurassic) ammonites from the United States and Alaska:

U.S. Geol. Survey Prof. Paper 249, 108 pp.
KEEN, A. M., 1958, Sea shells of tropical West America: Stanford, Calif., Stanford Univ.

Press, 624 pp.

KULP, J. L., 1961, Geologic time scale: Science, vol. 133, pp. 1105-1114.
LOEBUCH, A. R. 5 and TAPPAN, H., 1957, Correlation of the Gulf and Atlantic Coastal

Plain Paleocene and Lower Eocene formations by means of planktonic foraminifera:

J. Paleont., vol. 31, pp. 1109-1137.
MODEL, R., and MODEL, E., 1938, Die Lamberti-Schichten von Trockau in Ober-

franken: Jahrb. preuss. geol. Landesanstalt, vol. 58, pp. 631-665.
NATLAND, M. L., 1933, The temperature- and depth-distribution of some recent and

fossil foraminifera in the southern California region: Scripps Inst. Oceanog., B.,

Tech. Ser., vol. 3, pp. 225-230.
, 1957, Paleoecology of West Coast Tertiary sediments: Geol. Soc. Am., Mem.

67, vol. 2, pp. 543-572.

NELSON, C. A., 1962, Lower Cambrian-Precambrian succession, White-Inyo Moun-
tains, California: Geol. Soc. Am., B., vol. 73, pp. 139-144.
NEWELL, N. D., 1962, Paleontological gaps and geochronology: J. Paleont., vol. 36,

pp. 592-610.
OPPEL, A., 1856-1858, Die Juraformation Englands, Frankreichs, urid des siidwest-

lichen Deutschlands: Stuttgart, Ebner and Seubert, 857 pp.
D'ORBIGNY, A. D., 1842-1851 (1850-1860), Paleontologie fran$aise: terrains juras-

siques, cephalopodes: Paris, 642 pp.

CORRELATION BY FOSSILS 111

PAYNE, M. B., 1951, Type Moreno Formation and overlying Eocene strata on the west
side of the San Joaquin Valley, Fresno and Merced counties, California: California
Div. Mines, Spec. Rept. 9, 29 pp.

RANGHEARD, Y., and THEOBALD, N., 1961, Nouveaux gisements fossiliferes dans le Cal-
lovien moyen et superior de Besangon-Palente: Univ. Besangon Ann. Sci., 2me ser.,
Geol., Fasc. 15, pp. 3-7.

SCHENCK, H. G., and KEEN, A. M., 1936, Marine molluscan provinces of western North
America: Am. Phil. Soc. Pr., vol. 76, pp. 921-938.

SCHOELLHAMER, J. E., and KINNEY, D. M., 1953, Geology of portions of Tumey and
Panoche Hills, Fresno County, California: U.S. Geol. Survey Oil and Gas Inv. Map
128.

SCOTT, G., 1940, Paleoecological factors controlling the distribution and mode of life
of Cretaceous ammonoids in the Texas area: J. Paleont., vol. 14, pp. 299-323.

SMITH, W., 1815, A memoir to the map and delineation of the strata of England and
Wales, with part of Scotland: London, J. Gary, 51 pp.

, 1816, Strata identified by organized fossils: London, W. Arding, 32 pp.

SPATH, L. F., 1933, The evolution of the Cephalopoda: Biol. Rev., vol. 8, pp. 418-462.

STENO, N., 1669, De solido intra solidum naturaliter contento dissertationis prodromus:
Florence, 78 pp.

THEOBALD, N., 1957, Documents nouveaux sur I'extension des difierentes zones a am-
monites d'age callovien dans le Jura septentrional: Univ. Besangon Ann. Sci., 2me
ser., Gol., Fasc. 7, pp. 21-40.

ULRIGH, E. O., 1911, Revision of the Paleozoic systems: Geol. Soc. Am., B., vol. 22,
pp. 281-680.

WESTERMANN, G., 1954, Monographic der Otoitidae (Ammonoidea): Beihefte z. geol.
Jahrb. (Hannover), no. 15, 364 pp.

WOODRING, W. P., BRAMLETTE, M. N., and KEW, W. S. W., 1946, Geology and paleon-
tology of Palos Verdes Hills, CaUfornia: U.S. Geol. Survey, Prof. Paper 207, 145 pp.

DONALD B. McINTYRE

Pomona College

Precision and Resolution
in Geochronometry

There are probably compensating errors, and the resulting age determination
will be very useful to geologists z/, for instance, we could say that 9 times out
of 70 the age is x =t y million years. (Larsen, Keevil, and Harrisori,
1952, p. 1046)

Formerly it was the custom to distinguish between the so-called exact and
inexact sciences, an exact science being one which admitted, or perhaps we
should say whose practitioners claimed, absolute precision. Nowadays we are
more cautious, and most scientists are well aware that absolute precision can
be hoped for only when our measuring devices are being used far from their
limits. In other words, whether we are determining chemical composition,
stratigraphic thickness, or the age of a rock, it is common to find that if we
make replicate readings they will not be identical. Recognizing this, authors
often follow the report of a measurement by a plus-or-minus quantity which is
intended to indicate, in some manner, the degree of confidence which is to be
placed in the value given. Unfortunately, the reader is not always informed
what the basis for the plus-or-minus quantity is and what confidence is to be
assigned to it, and to provide the figure in such cases is obviously an empty
gesture.

In this essay the degree of reproducibility is referred to as precision; this is
not to be confused with accuracy which is the approach to a hypothetical "true"
value. We will assume that accuracy (or bias) is dependent upon systematic
errors, such as faulty adjustment of instruments or the use of incorrect constants,
whereas precision is determined by numerous small and independent random
errors. * Discussion of the existence and possible magnitude of systematic errors
and of gross errors lies beyond the scope of this essay.

1 The implied hypothesis of normal distribution may be questioned, but until we are
given data demonstrating how the distribution departs from the normal, we have no
satisfactory alternative but to accept it.

112

PRECISION AND RESOLUTION IN GEOCHRONOMETRY 113

Geology is rapidly becoming quantitative and it seems worth while to draw
attention to the importance of presenting data so that the precision of measure-
ment is clear, for it is this precision that determines the resolving power and
hence, in large measure, the utility of the method. Although the topic is rele-
vant to any measurements, geochronometry happens to present convenient
examples. It is not the writer's purpose, however, to review the relative merits
of different methods of age dating. The two methods which are discussed,
potassium-argon and lead-alpha, were chosen as illustrations because they have
yielded the majority of dates on the Mesozoic batholithic rocks of western
North America. The writer is not a practitioner of geochronometry; his quali-
fications are those of most members of the geological fraternity, for we have
all been addressed on this subject through our professional literature, and it is
our responsibility to draw conclusions from the information published.

In 1960 this essay was sent to many who are active in geochronometry, and
the writer is grateful for the replies and comments which he has received. In
the meantime considerable advances have been made in isotopic studies, and
some doubt has been expressed regarding the accuracy of the lead-alpha
method. Nevertheless, because the purpose of the essay is to encourage other
readers to draw their own conclusions from any data presented to them, it
does not seem necessary to substitute other examples for those originally
included.

Since Rutherford's suggestion, in 1905, that the accumulation of decay
products in radioactive minerals might provide a means of measuring geological
time, great numbers of minerals and rocks have been "dated." Indeed, the
rate of output of new "dates" is now so high that it is difficult for evaluation
to keep abreast of production. Many geologists have all but given up the
attempt and are ready to accept without question any age correlation or dis-
tinction made by such methods.

A date quoted for a rock is no better than its precision; hence readers are
entitled to expect every author to pass on, in unambiguous terms, his best
quantitative evaluation of this date. Because the precision of the age is a func-
tion of the precisions of each variable entering into the computation, we must
begin by analyzing the random errors in each of these variables.

Precision of the Lead-Alpha Method

"A mineral to be usable for determining the age of rocks by this method
should have only radiogenic lead, should have lost no lead, should have only
primary radioactive elements, and should have lost none of them." (Larsen
ft al., 1952, p. 1046) The ionic radius of Pb 2+ (1.32 A) is very close to that of
K+ (1.33) and very different from that of Zr 4 + (0.87); hence it is reasonable
to suppose that primary lead will be preferentially captured (Goldschmidt,

114

DONALD B. McINTYRE

TABLE 1

TWENTY-FIVE REPLICATE DETERMINATIONS OF PB IN SAMPLE

CONTAINING 30 PPM
ANALYSES BY C. L. WARING AND H. WORTHING (1953)

Observation number

Pb, ppm

* 2 *

i

x*

1

31

961

2

30

900

3

33

1089

4

30

900

B

34

.

B

30

.

.

29

.

30

28

28

28

29

30

30

32

28

31

.

30

.

.

35

.

.

32

m

21

30

900

22

29

841

23

30

900

24

31

961

25

32

1024

n 25

Et-i x, = 760

Z?-i ^* = 23184

r 5 * r Y^ ft f^T Vl

3.33,

n( 1)L '^ vi-V J -

1_

J = - r = 30.4,

^ = 1.82 ppm or 1.82 X 100/30.4 = 6.0%.

* In practice the figures in column i and x* are not recorded, for only the totals are
used and these are obtained on a desk calculator as by-products of entering the values *,-.

PRECISION AND RESOLUTION IN GEOCHRONOMETRY 115

1944; Gottfried, et aL, 1959) in orthoclase, whereas any lead found in zircon
is likely to be radiogenic from a U 44 " (1.05) or Th 4+ (1.10) parent. Isotopic
analyses confirm that the proportion of primary lead in zircon is often small
(Gottfried, et aL, 1959, Table 14), although there are notable exceptions. On
the assumption that the mineral has had a history which makes the method a
valid one, it is necessary to determine (a) the radioactivity, by means of an
alpha counter, (b) the thorium/uranium ratio, which for accessory zircons is
usually considered known from earlier work, and (c) the amount of lead by
spectrographic analysis.

Precision of lead determination. In the application of the method to zircon
concentrates from Mesozoic rocks, it is necessary to determine the lead content
when it ranges from 1.8-180 ppm (Larsen, et aL, 1958). In order to estimate
the precision of the lead determinations, Waring and Worthing (1953) made
25 replicate measurements (reproduced in Table 1) on a sample known to
contain 30 ppm lead in a silica base. We consider these 25 measurements as a
sample (size n = 25) which is representative of the population of all possible
measurements. Each individual measurement is termed a variate, x^ i = 1,
. . . , n; and the statistics of the sample are:

I

the mean, x = - Yj * = 30.4 ppm,
n ~r.

2 1 ^~* / \ 2 * F ^ 2 /^ \2 I

the variance, s = T >, (*i x) = -7 TT n^x (2*x) \,

n - 1 J n(n - 1)L J

and the standard deviation, s = 1.82 ppm or 6.0%. 2

These definitions are in forms suitable for use with a desk calculator on which
the sums of x* and of x can be accumulated simultaneously. They give the
best estimates of the parameters* of the population from which the sample has
been drawn. From the table prepared by Eleanor G. Crow (in Crow, et aL, 1 960,
Table 8), the standard deviation of the population is a = 4.9-7.9% with
90% confidence or 4.7-8.4% with 95% confidence. This gives a meaningful
measure of the precision of the method.

2 WQs/x is the coefficient of variation. It is often denoted by C, but here we will use
simply j%, and, by analogy, <r%. The transformation is justified if s is proportional to *;
in some of the examples this may not be the case.

8 It is convenient to use italic letters, such as x and s, for the statistics of a sample and
to use Greek letters, such as /* and <r, for the parameters of a population. In practice
we are interested in the statistics only as estimates of the parameters.

116 DONALD B. McINTYRE

In 1959, Gottfried, JafTe, and Senftle (1959, Table 8) presented a table
showing how the precision of the lead determination (at confidence level of
90^) varies with the amount of lead present. If the percentage standard
deviation is plotted on semilog paper against the geometric mid-points of the
lead ranges given by these authors, we may estimate 4 the precision of all lead
determinations on zircons published in U. S. Geol. Survey B. 1070-B (Larsen,
et a/., 1958, Tables 5-11). The mean value of the standard deviations estimated
in this way is 10%. However, many of the lead-alpha dates on rocks of the
Mesozoic batholiths were determined on the basis of average values of several
lead determinations. In these cases the precision is that of the mean,
cr = CV/X/H, where n is the number of replicate determinations included in
the mean. The mean value of all the standard deviations estimated in this
manner is 8.8%.

An independent evaluation of the precision of the lead determinations can
be obtained from an analysis of the duplicate results themselves. A pooled
estimate of the mean precision of the individual lead determinations, i.e., a
mean value giving increased weight to larger values of j and of n was computed
as s p where A" is the number of samples on which replicate determinations were
made and

= Zf-i ( - D4 =

Hence we estimate (7 = 9.4-13.6% with 90% confidence. This is consistent
with the value of 10% obtained above on the basis of the application of Gott-
fried, Jaffe and Senftle's Table 8 (1959) to the data in U. S. Geol. Survey B.
1070-B, Tables 5-11. We therefore adopts = 9-13% as a tentative 5 measure
of the precision of lead determinations for the evaluation of the dares obtained
on the Mesozoic batholithic rocks of western North America. That this value
is reasonable is further borne out by the fact that replicate analyses of lead in
accessory zircons from eastern Massachusetts granites (Webber, et al., 1956,
Table 2) yield a pooled estimate of <r = 9.2-12.0% with 90% confidence
(see Table 2).

4 The interpretation of Table 8 of Gottfried, et d. (1959) is ambiguous, but the writer
believes that any other method of plotting will give higher values for <r. He has assumed
that 1.645<r = precision at 90% confidence.

5 Since this essay was written, the work of Rose and Stern has suggested that a better
precision (<6%) may be obtained; but whether the values published in U. S. Geol.
Survey B. 1070-B justify this higher confidence is another matter. This illustrates one
of the difficulties faced by the reader who wishes to evaluate published dates.

PRECISION AND RESOLUTION IN GEOCHRONOMETRY

TABLE 2

117

ANALYSIS OF REPLICATE LEAD DETERMINATIONS ON ACCESSORY ZIRCONS

FROM EASTERN MASSACHUSETTS GRANITES
BASED ON DATA IN WEBBER, HURLEY, AND FAIRBAIRN, 1956

n

Pb, ppm

s, ppm

s%

n

Pb, ppm

s, ppm

*%

6

46.5

2.6

5.6

6

68.3

8.6

12.6

6

21.3

1.0

4.9

3

73.0

2.0

2.7

6

19.2

1.9

10.1

4

70.2

3.3

4.7

3

141.0

20.4

14.5

2

89.5

0.7

0.8

6

85.8

2.7

3.2

3

77.0

5.0

6.5

2

58.0

18.4

31.7

3

117.0

6.1

5.2

6

37.2

6.4

17.3

3

116.3

14.8

12.7

4

28.0

0.8

2.9

5

81.8

10.1

12.3

4

46.8

2.1

4.4

3

75.3

3.8

5.0

3

76.3

9.7

12.7

5

65.6

7.4

11.3

3

56.3

7.5

13.3

3

55.0

5.3

9.6

7

71.4

8.7

12.2

3

66.0

1.7

2.6

2>

1) = 75

V (* 1 > -2
2^(n i)sr

8111.5

8111.5
75

108.15

Hence s p = 10.4%, and<7 = 9.2-12.0% with 90% confidence.

Precision of Th/U ratio. Because the decay constants of thorium and uranium
differ, it is necessary to know the Th/U ratio in a mineral which is to be dated
by the lead-alpha method. Most of the dates obtained by Larsen and his
associates on the batholithic rocks of western North America were based on
accessory zircon, and it is this case which is treated here. The available data
on the Th/U ratio in accessory zircon have been summarized by Gottfried,
Jaffe, and Senftle (1959, Table 1). It is interesting to note the following points:
(a) none of the data were published prior to 1957, (b) only 13 "localities" are
represented, (c) the Mesozoic batholithic rocks of western North America are
not included, (d) a considerable variety of analytical methods were employed
providing accuracies said to vary from 1% to =b20%, (e) at only seven of
the "localities" were measurements made on more than one sample, and hence
these are the only ones for which a* can be estimated, and (f ) the number of
samples, the range, and the average are given for these seven localities, but s is
not given.

Because complete data are not available for calculation of j, we have to
estimate it from the range (see Crow, et a/., 1960, Table 12). In four cases it is

118

DONALD B. McINTYRE

TABLE 3

Th/U RATIOS IN ACCESSORY ZIRCONS

Number of samples. Range, Mean value. Estimate of
it, w Th/U-*, *

2 90% confidence
*. limits for <r<

1

0.8

22

2.2

1.2

0.59

7.31

1

1.2

21

2.3

0.9

0.62

7.68

0.37-0.62*

24
10

0.8
0.2

0.5
0.2

0.21
0.07

1.01
0.04

0.16-0.26f
0.06-0.14f

1

0.3

9

2.4

1.2

0.81

5.25

0.58-1.38J

1

0.4

1

0.5

4

0.32

0.3

0.16

0.08

2

0.24

0.4

0.21

0.04

1

1.1

!> = 98, where

HI > 1, !> = 92,

and .V =* 7.

Pooled mean for Th/U - 0.78

21.41

Pooled estimate s% =
Total range = 2.7;

21 41
'

; hence J P = 0.50, andcr = 0.44-0.57 \vith 90% confidence
98; hence standard deviation can be estimated as 0.54

Columns 1 through 3 of this table computed from Gottfried, et al. (1959), Table 1.

* Figures computed from the data of H. Holland (in Gottfried, et al., 1959).
f Figures computed from Hurley and Fairbairn (1957).

* Figures computed from Lyons, tt a/. (1957).

possible to refer to the original data, and for these the writer has computed
90^-confidence limits for cr (see Table 3). In each case the values of standard
deviation computed from the range fall within these limits. The pooled mean
for Th/U is 0.78 and the unweighted mean is 0.69. These figures do not accord
very closely with the claim that: "At present it seems reasonable to assign a
thorium-uranium ratio of 1 for the average accessory zircon." (Gottfried,
et aL, 1959, p. 12) If the wrong ratio has been used, it is quite possible that this
would produce a systematic error only, and that the relative ages of the rocks
of a single province might not be affected. However, it is obvious from Table 3
that even within a single province the Th/U ratio in accessory zircon can vary

PRECISION AND RESOLUTION IN GEOGHROXOMETRY 119

considerably. Thus, the variability of Th/U must be estimated, for it will
affect the precision of the date. Because 2.7 is the total range of all 98 values
on which Table 3 is based, we can estimate the standard deviation as 0.54.
This agrees with the pooled estimate of the standard deviation, s = 0.50,
giving a- = 0.44-0.57 with 90% confidence.

Now, the Th/U ratio enters into the age computation by way of the factor c ,
where

_ 2632 + 624 Th/U
*~~ 1 +0.312 Th/U "

(See Gottfried, et a/., 1959, Fig. 1.) We have determined that the mean value
of Th/U is about 0.78 and that the standard deviation of the possible values
for this ratio is likely to be about 0.50. Hence a Th U range of <7 would be
0.28-1.28, giving a range of 2581 through 2452 for c. Hence or for factor c can
be estimated as about 2.6%, or approximately 3%.

Precision of alpha measurement. The radioactivity is measured by the ion-
chamber method and is given as the number of alpha counts per milligram
per hour. Gottfried, Jaffe, and Senftle (1959, p. 22) report: "... the activity
determined by counting methods agrees on the average within 5 percent with
the alpha activity calculated from the thorium and uranium content ... In
general, the interlaboratory checks agree within 5 percent. From these checks,
and also from the alpha-counting experience in this and other laboratories,
the alpha-counting data probably are accurate to within 5 percent.* 5 Lyons,
et aL, (1957, p. 528) have recorded their precision in the statement: "Duplicate
measurements of alpha activity on the same sample are reproducible to within
5% of the measured value. 95 Webber, Hurley, and Fairbairn (1956, Table 3)
have given figures for "standard-deviation precision error based on replicate
alpha counts 55 ; their data as published yield a mean value, 5%, of 7.8 with
cr = 6.4-10.1% with 90% confidence. Our conclusion must therefore be
vague, with <r for alpha determination surely somewhere in the range from
3-10%.

Resulting precision of age determination. If the age is less than 200 million years
(m. y.), it is determined from the equation t = (c - Pb)/a, where t is the age
in millions of years, c is the factor depending upon the Th/U ratio, Pb is the
lead content in parts per million, and a is the radioactivity in alpha counts
per milligram per hour. If the partial errors in / due to given errors in c, Pb,
and a are E C9 pb and E a respectively, then the total error in / is E t = E c +
Efb + * In practice we know only the probability distributions of these
partial errors and not their actual values. Thus, although it is true that the
total error is the sum of the partial errors, the parameters of the probability

120 DONALD B. McINTYRE

distribution of t cannot be obtained by summation of the parameters of the
probability distributions of the individual variates: i.e., <r t 7* <r c + 0"Pb + 0"*-
When we make an age determination, what we do in effect is to pick at random
the magnitudes of the errors in c, Pb, and a and then add their effects.
Assuming that there is no correlation between E c , Epb, and a , i e., that there
is no reason to expect a particularly large error in c to occur along with a
particularly large error in Pb or in a, then we have of = <r? + <Tpb + <r. 6
(If this assumption is not correct then the correlation coefficients must be
determined and used to derive of.) As a consequence of this relationship,
if <rpb is considerably greater than a c or <r a , as may be the case, then ff t will be
only a little larger than o-pb- Using the best estimates which we have derived
above, we obtain

(Tp b - 9-13%, <r a ~ 3-10%, (T C ~ 3%.

Hence, the predicted precision of dates on the Mesozoic batholithic rocks by
the lead-alpha method is a ~ 10.0-16.7%, or about 9.8-13.6% if allowance
is made for the replicate determinations of lead.

The observed precisions (based on Larsen, et at., 1958, Table 12) are as
follows:

s% 90%-confidence limits

35

Mexico and Southern California

10.3

8.6-12.9

15

Sierra Nevada

10.8

8.3-15.7

16 7

Idaho

11.1

8.6-16.0

16

Coast Range

12.4

9.6-17.8

82

All 82 specimens taken together

11.2

9.9-12.9

We conclude that (a) the observed and predicted precisions are the same,
within the limits imposed by the data available to us, and (b) all 82 specimens
from the Mesozoic batholiths of western North America have the same age,
within the limits of the method.

In their work on the accessory zircon from the eastern Massachusetts gran-
ites, Webber, Hurley, and Fairbairn (1956, Table 4) state "standard deviation
precision errors" for the ages determined. From their data we estimate <r =
7.7-12.3% with 90% confidence.

6 One reason for the common use of standard deviations is that their units can be those
of the original observations. But in practice there are advantages when variances are
employed; the additivity referred to here is one of these.

7 Incorrectly given in loc. ciL

PRECISION AND RESOLUTION IN GEOCHRONOMETRY 121

The Potassium-Argon Method

The potassium-argon method is based on the assumption that all potassium
in nature contains the same percentage of radioactive K 40 and, further, that
the mineral to be dated has neither gained nor lost any potassium or any
radiogenic A 40 since its crystallization. The method and the history of its
development have been outlined by Carr and Kulp (1957) and by Lipson
(1958), and possible sources of systematic error have been discussed by Curtis,
Evernden and Lipson (1958), Curtis and Reynolds (1958), and by Centner
and Kley (1958).

The source of the greatest practical difficulty which is encountered in this
method is that 99.6% of atmospheric argon is A 40 , identical with the minute
amount of radiogenic argon which has to be extracted from the mineral and
measured. Hence the A 38 and A 36 , together forming only 0.4% of atmospheric
argon, provide the only means by which the amount of contamination by
atmospheric argon can be determined. In practice a "spike 55 of pure A 38 is
added to the sample as an isotopic tracer and the A 40 / A 38 and A 38 / A 36
ratios are each measured on the mass spectrometer.

Total potassium is usually determined by flame photometry and the amount
of K 40 is then computed using 0.000119 as the atom fraction of K 40 in potas-
sium and 39.100 as the atomic weight.

Precision of radiogenic A 40 /K 40 ratio. The writer has found great difficulty
in discovering just what some authors mean in their statements regarding
precision. Some use standard deviation or probable error explicitly (cr =
1.4826 X PE), but others refer vaguely to "maximum error," do not dis-
tinguish clearly between precision and accuracy, or make statements such as
"duplicate measurements are usually reproducible to within /?%." Frequently
even the number of determinations and the range are not stated. In many
instances it is obvious that the authors had data from which useful measures
of the precision could have been obtained. By not sharing them with their
readers, they have diminished the value of their work, for confident inferences
cannot be drawn unless the precision is known.

An explicit statement regarding precision has been given by Lipson (1958,
p. 144):

The A 40 /K. 40 ratio may now be computed with a probable error of from
5 percent to 7 percent, depending upon the percentage of atmospheric argon
in the sample. The total probable error arises from a probable error of
3 percent in the potassium determination and a probable error of 4 percent
in the radiogenic argon determination exclusive of error in the atmospheric
argon correction. The error in the determination of the A 36 /A 40 abundance
ratio is taken to be 3 percent, but the effect of this error on the radiogenic

122 DONALD B. McINTYRE

argon determination varies with the percentage of atmospheric argon. If
we let

e = the percentage error in A 36 /A 40 ,
/ = percentage of atmospheric argon in the sample,
E = the percentage error in the radiogenic argon due to the error in
A 36 /A 40 ,

then

Applying this equation by using the percentages of atmospheric argon which
are tabulated for each sample, the total probable error can be calculated.

If the percentage of atmospheric argon rises above 50, then, as a consequence
of the relationship cited, the error in A? 'K 40 increases very rapidly. Unfor-
tunately, the percentage of atmospheric argon is not always stated, and not
infrequently, it is impossible to tell whether this error has been included in
the precision given for A 40 ' K 40 . Centner, Jensen, and Mehnert (1954) report
10-1 5% of atmospheric argon in the determination of a feldspar from a pegma-
tite in the Black Forest. Such a percentage gives an error of only or = 0.5-0.8%
in the A 40 , K 40 ratio. On the other hand, for the 10 New Zealand glauconites
dated by Lipson (1958), the atmospheric argon varied from 37-68%, cor-
responding to a range in precision for A 40 /K 40 of a = 2.6-9.5%. In the
dating of 10 granitic rocks from Yosemite (Curtis, Evernden, and Lipson,
1958), the range in percentage of atmospheric argon was 8.5-63.2, which
corresponds to a range of precision for the A 40 /K 40 ratio of <r = 0.4-7.6%.
Yet, referring to the Yosemite analyses, Evernden, Curtis, and Lipson (1957,
pp. 2120-2121) wrote:

The major error in our technique is concerned with the absolute calibra-
tion of the argon-38 spikes used for isotope dilution. This calibration error
may give rise to a probable error of 2 or 3 per cent. However, it must be
emphasized that, with the spike preparation procedure now being used, the
argon-38 spike quantities are known relative to one another to less than
1 per cent. Thus it is believed, if the error introduced by uncertainty in the
absolute-spike calibration constant is negligible, that the analytical technique
has an inherent probable error of less than 1 .5 per cent. Repeat runs on split
fractions of biotite concentrates substantiate this conclusion.

No data on these reproducibility tests are given. The Yosemite dates are
stated as =tl-2%. With reference to the same rocks, Curtis, Evernden, and
Lipson (1958) state that, for the potassium analyses, if duplicate determina-
tions do not check within 1.5%, they are rejected. However, this does not

PRECISION AND RESOLUTION IN GEOCHROXOMETRY 123

tell us what the precision of a potassium analysis is. Precision can be improved
only by perfecting the method or increasing the number of replicate determina-
tions; it cannot be improved by arbitrary rejection of supposedly extreme
values (See for example Crow, et al., 1960, Table 16).

In 1954, Wasserburg (in Paul, 1954, p. 342) wrote: "Since argon 40 : potas-
sium 40 ratios can be determined to within 4 percent, it should be quite possible
to resolve Paleozoic intrusions separated by one geological period." If 4%
is the probable error, thencr = 6%. In 1956, Folinsbee, Lipson, and Reynolds
(1956, Table I) gave the A 40 /K 40 ratio as 5%, although they did not show
how the figure was derived. Presumably a = 7.5%. Aldrich, et aL, (1956,
p. 217) make the following statement: "From the reproducibility of the results
by the two methods, it is believed that the error in the potassium analyses is
5 percent or less. From the determinations of argon in air and minerals, the
accuracy of the argon determinations is believed to be 3 percent. The error in
the ratio, A 40 /K 40 , is believed to be less than 6 percent." Presumably cr < 9%.
Carr and Kulp (1957, p. 777) report: "The numerous repeat runs on the same
powder show that the reproducibility in the A 40 determination lies in the
range of 1-3 percent in most cases. Since this is also the range of variation
of the replicate potassium analyses, the analytical error in the A 40 /K 40 ratio
probably does not exceed 5 percent."

With reference to the dating of basement rocks from the Black Forest,
Centner and Kley (1958, p. 103) give the following values for the nonsystematic
errors: potassium analysis 1-2%; argon determination 2%; atmospheric argon
correction 1-2%. If these figures are probable errors, then <7A 40 /K 40 is in
the range 3.6-5.1%. These authors also state (p. 103) that the individual
measurements can fluctuate through a maximum of 10% on a relative time
scale. If by "maximum" is meant "with 95% probability," then we can con-
clude that <7A?/K 40 is about 5.1%, whereas if "with 99% probability" is
meant, then a turns out to be about 3.9%.

These examples are sufficient to indicate how difficult it is to determine from
the literature what standard deviation should be accepted for the ratio of
radiogenic A 40 /K 40 on which the age of a rock may depend. The obvious
conclusion is that authors should be sure that their readers know exactly what
is meant by the precision stated, what it is based on, and what confidence
should be attached to it. The writer suggests that until this important question
is clarified the most reasonable theoretical precisions to adopt for the purposes
of evaluating dates are as follows:

Potassium analysis 3-6

Total argon determination 3-6

Atmospheric argon correction 0.5-8

Spike calibration 1.5

124 DONALD B. McINTYRE

Assuming that there is no correlation between these errors, we find that the
precision is in the range or = 4.5-11. 8^. It is unfortunate that this important
parameter is not better known.

Resulting precision of potassium-argon age. K 40 decays into Ca 40 with a constant
denoted by X& as well as into A 40 with a constant XK (.sometimes written X e ).
The ratio between these two constants, XK X$, is designated /?, the branching
ratio. The total rate of decay of K 40 is given by X = XK + X$. The conver-
sion of K 40 into Ca 40 is accompanied by emission of /3-rays (electrons), which
are easily detected; hence \& is comparatively well known. On the other hand,
the conversion of K 40 into A 40 is effected within the nucleus by transformation
of a proton into a neutron by capture of one of the electrons from the A"-shell.
In the consequent rearrangement of the remaining electrons, which is necessary
in order that the vacancy in the A"-shell be filled, there is emission of low-
energy x-rays, the characteristic A-radiation of potassium. Thus XK can be
determined only indirecdy, and the uncertainty in the value of this constant
is one of the major sources of systematic error in the method at the present
time. Centner and Kiev (1958) claim that whereas X, the total disintegration
constant of K 40 , is known to within 3-4%, the branching ratio, R, may be in
error by as much as lO^.

However, any errors in the constants will not affect relative age measure-
ments, provided that the same values are used for all determinations. Because
we take as an example the granitic rocks of Yosemite, reported on by Evernden,
Curtis, and Lipson (1957) and by Curtis, Evernden, and Lipson (1958), we
adopt the values used by these authors:

X0 = 0.472 X lO^yr" 1 ,

X K = 0.557 X lO-^yr" 1 ,
R = 0.118,
X = 0.528 X 10" 9 yr" 1 .

Now the equation used in calculating the age of a potassium-bearing mineral
is

T= -InTl + 1 + *1

where T is the age in years. Letting x = A 40 /K 40 and t = age in millions of
years, we have dt/dx = 2.12/(0.118 + 1.118*) million years per 10~ 3 unit of
A 40 /K 40 . Even after 100 m.y., A 40 /K 40 is still less than 0.006, and thus the
gradient changes only very slowly with increasing age. A 1% error in A 40 /K 40
produces about the same error in the age if the rock is less than some 200 m.y.
old. If the rock is older than about 800 m.y., then the error produced in the

PRECISION AND RESOLUTION IN GEOCHROXOMETRY 125

age is appreciably less than the error in A 40 K 40 . Thus, unless we are dealing
with Precambrian rocks, or the precision is kno\\n \\ith great confidence, it
can be assumed that a given percentage error in A 40 K 40 will produce about
the same percentage error in the age.

If replicate age determinations are made on one or more rocks, then the
results can yield a value for the "observed precision" of the age, which can be
compared with the "theoretical precision" determined from consideration of
the analytical techniques. It appears that the data on a suite of rocks from the
Black Forest (Mehnert, 1958) may be used in this way, but it is unfortunate
that experiments have not been undertaken with this goal specifically in view.
Fourteen specimens were collected from the basement rocks of the Black Forest,
and from some of these, feldspars and micas were separated so that 19 prepara-
tions were available for dating, and on these, a total of 66 determinations were
made. For the present purposes, we are concerned solely with the degree of
reproducibility. An analysis of Mehnert's data is given in Table 4. For each
preparation we compute the standard deviation as a percentage of the mean
so that the effect of variation in the means may be eliminated. A pooled esti-
mate of the standard deviation is s p = 10.8%, with a t = 9.0-13.5% with
95% confidence, which is to be compared with the predicted precision of
a t = 4.5-12%. If the analyses given by Mehnert are truly replicate, it would
seem reasonable to suppose that the precision of age determinations by the
potassium-argon method could be as much as a t = 10%.

Statistical Inference

Sampling distributions. If a random sample of size n is drawn from a normal
population with mean p x and standard deviation <r x , then, in general, the mean
of the sample is x 7* p x . But the number of possible samples of size n which
could be drawn from the population may be large, and the assemblage of
all these possible samples forms a second-order population whose parameters
are related to those of the original population as follows:

the mean value of x = /** = p x ,
the standard deviation of the sampling distribution
(standard error of *) = G = cr x /\/.

For example, suppose we draw a sample of size 10 from a population of 100 indi-
viduals with ps = 12 and v x = 4. Our sample is one out of 100!/(10! X 90!)
or more than 10 13 possible samples. The mean of this sampling distribution
(population of samples) is fig = 12, and its standard deviation (standard
error) is cr $ = 4/VTO = 1.26. Similarly, if the standard deviation s is deter-

126 DOXALD B. McINTYRE

TABLE 4

REPLICATE AGE DETERMINATIONS ON 19 ROCKS AND MINERALS FROM

THE BLACK FOREST BY THE POTASSIUM-ARGON METHOD

BASED ON DATA PUBLISHED BY MEHNERT, 1958

Number of
determinations

Mean age

s%

5

249

2.0

2

254

0.3

3

243

2.4

3

268

2.6

5

253

4.5

3

269

3.5

4

287

1.6

5

241

16.1

3

273

0.9

2

306

0.1

3

303

3.8

3

280

2.7

2

260

1.1

6

256

25.5

5

278

15.1

3

344

2.4

3

304

1.3

3

283

1.5

3

310

6.6

Hence jp = 10.8% and<T =

9.0-13.5% with

95% confidence

mined for all possible samples of size n, then the standard deviation (standard
error) of this sampling distribution is <r 8 = tr x /-\/2n. For the example given,
we would expect the standard deviations of samples of size 10 to have a distri-
bution with a mean of /z = <r x = 4 and a standard deviation of

<r g = 4/V20 = 0.9.

Suppose we have two populations defined by parameters AC* and <r x , and /i y
and <r tf , respectively, and we draw a pair of variates, one from each population,
and take the difference (x y). If we continued to draw pairs until we had
determined all possible values of (x 7), then we would have a new second-
order population (sampling distribution) with parameters

= MX Pt/ and <rL. v = a* + cri

PRECISION AND RESOLUTION IN GEOCHRONOMETRY

127

.960*

FIG. 1 . If the precision of an analytical method is given by (7, \ve have 95% confidence
that a single determination will not differ from the mean, j&o> by more than 1.960(7.

provided that there is no correlation between members of the pairs. Thus if
&x = &y = 0*3 then ffx-y = (7\/2.

If we are concerned with differences between means instead of between vari-
ates, then we have

M2 y ~~~ M2 Mys

2 2

= ,'(1 + 1),
\BI n 2 /

2cr 2

if (7a; = <7 y = (7,

j if, in addition, ?zi = 2 = -

n

Critical value. Suppose that the precision of chemical analysis of an element
is already known and is given by <r, and further, let us assume that the distribu-
tion of numerous replicate analyses would yield a normal curve with mean /z
and standard deviation cr (Fig. 1). Because we find, from tables of areas under
the normal curve, that 95% of the distribution lies between 1.960(7 and
+1.960(7, the probability is only 0.05 that a single determination will fall
outside the range JLI O 1.960(7. Accordingly, with 95% confidence, we reject
the hypothesis that a determination outside this range belongs to this popu-
lation. Now at this confidence level, five times out of one hundred, we will
make a mistake and reject a measurement erroneously, claiming that it does
not belong to the population when in fact it does, as we would discover if
we were to make more determinations. The error of rejecting the null hy-

128 DONALD B. McINTYRE

pothesis when in fact it is true is called a type I, or producer's, error, for when
it is made, a perfectly "good" sample is rejected. In our example, the chance
of making a type I error is a. = 0.05; this is called the significance level. Now
it may be that the cost of producing each measurement is so high, or other
consequences of making this error are such that we cannot afford to reject
such a high percentage of good determinations. If this is so, we can decrease
the critical rejection area, say by setting a = 0.01, with 0.005 at each tail of
the distribution, or even less. From tables of percentiles of the normal distribu-
tion we find the following:

Probability of making Critical value for

Type I error \x /io|

0.10

1.645<7/V*

0.05

1.960<r/V

0.01

2.576<7/Vn

0.001

3.291o-/V

We can also make use of these relations in the following way: if we have deter-
mined x from a sample of size n and we already know <r, then we can state
confidence limits for the mean, ju 05 of the population to which x belongs. For
example, with 95^ confidence, we can say 8 /i = * 1.960or/\/fl- Alter-
natively, the sample size required to specify ju to the range A with 95%
confidence is n = (1.960er/A) 2 .

Resolution. We have seen that we can always lessen the probability of making
a type I error (i.e., claiming that the null hypothesis is false when it is actually
true) merely by decreasing the rejection area a. This is true even when the
sample size is as small as two. Whatever value of a we decide upon, we calcu-
late the critical value and agree to reject the null hypothesis if \x /*o| is
greater than this. Now, clearly the smaller we have made a (to avoid type I
errors), the more likely it will be that we accept the null hypothesis when it is
false. This would be a type II error, and the probability of making it is given
by (Fig. 2). We see that, although [^i /* | is greater than our critical
value, the probability of overlooking this difference is ft. This error is often
called a consumer's error, for it will result in the acceptance of measurements
which should be rejected. Our confidence in avoiding this error is 1 and
is the power of the test. Clearly 0, and hence the power, depends upon |/ii j* |
<r, n, and a. If four of these variables are given, then the fifth is fixed. Let us
suppose that <r and n are given and a has been assigned the value 0.05, so that

8 If a were not known, we would use the ^-distribution (see. for example, Crow et al.,
1960, p. 47).

PRECISION AND RESOLUTION IN GEOCHRONOMETRY

129

-30

70

Million years

FIG. 2. a is the probability that we will reject, in error, the null hypothesis that a
sample mean belongs to population JLIQ because it is greater than JLIQ + h or less than
MO h. (1 a) 100% is our confidence that we will not make this error, ft is the
probability that we will accept, in error, the null hypothesis that a sample mean belongs
to population JUQ because it is less than JUQ + h when an alternative hypothesis, namely,
that the sample mean belongs to population MI, is true. (1 )100% is our confidence
that we will avoid this error. The diagram is drawn with a. = 0.05 and ft = 0.2.

the critical value for rejection is k\<r$ = 1.960cr/\/. It is clear that ft must be
much larger than a. unless |/*i /x | is made extreme. Let us suppose that ft
is not to exceed 0.2; then, from normal curve tables.

Hence

0.8420-
VS '

2.8020-

MO|

for a = 0.05 and ft = 0.2. Similarly,

~ MO| =

3.242(7
Vn

for a. = 0.05 and ft = 0.1. These values of \m /o| are called the resolution 9
differences. The values of a and ft that we use will depend upon the penalty
attached to these errors in each particular case and on the economics involved.
In what follows, we adopt the values a = 0.05 and ft = 0.2, thus indicating

9 The analogy with the resolving power of an optical system will be obvious.

130 DONALD B. McINTYRE

that we expect to make a type I error (i.e., to claim a difference jjui Moi >
when the difference is actually zero] once in 20 times and a type II error
(i.e., fail to detect an actual difference |MI Mo' > Q) once i n 5 times. It
does not seem worthwhile to make statements which have lower degrees of
confidence, but on the other hand if these risks seem too great, then they can
be reduced: the price is an increased value for the minimum difference, R,
which can be resolved, and this penalty can be avoided only by improving the
precision of the individual measurements (thereby reducing 0") or increasing n,
the size of the samples used in determining x.

Summarizing these conclusions we can make the following statements:

(1) With 80^ confidence, we can claim that if JMI MO| exceeds 2.802(7 ^/H,
we will detect this at the 0.05-significance level.

(2) With 95^ confidence, we can claim that if |3c MO! exceeds 1.9600yS/J2,
then MI ^ Mo-

(3) The number of variates to be included in each sample in order to establish
a given degree of resolution = (2.8020", R) 2 .

In geochronometry we commonly have data on two samples and wish to
know whether the populations from which they were drawn have a common
mean. For instance, two samples are analyzed for potassium: the number of
replicate determinations made on the first is HI, and on the second n 2 ; the
respective means are x and J. Let us suppose that the precision for the method
is already established so that ff x = <r y = a which is known. Then we have
the following results.

For ,MZ M ; :

/I 1\ 1/2 0-

Resolution, R = 2.8020- ( + j = 3.963 -^ if ni = n 2 = ;

for \x Jj:

f\ IN 1 ' 2 0-

Critical value = 1.9600- ( h I = 2.772 -7=. if HI = n 2 = n:

vzi n 2 / Vn

and

.2

to establish a given resolving power.

These values used in testing the hypothesis (MI "" Ma| = are greater by the
factor \/2 than the comparable values for the hypothesis |MI MO| = 0. The
reason is that whereas MO was given, neither MI nor M2 is known.

We may therefore state criteria for decision-making as follows. If MO is
known and <r x = <r v = v is known but MI and M2 are unknown, then (1) with
100(1 0)% confidence, we can claim that if a difference exceeds the reso-

PRECISION AND RESOLUTION IN GEOCHRONOMETRY 131

lution, we will detect this at the a significance level:

100(1 0)% Difference Resolution a

80

IMI MO|

2.802cr 'v^

0.05

80

IMI M2J

3.963(r/V^

0.05

90

IMI ~~ MO!

3.858<r/V^

0.01

90

IMI M2J

5.456cr/\/7z

0.01

(2) with 100(1 a)% confidence we can claim that if a difference exceeds
the critical value for rejection, then m 9* a:

100(1 a)% Difference Critical value a

95

|* - MO| 1.960(7/V^

MO

95

/I 1\ 1/2
\x y\ 2.772cr/Vn or 1.960<r ( + )

\fll T12/

99

\x jitol 2.576cr/Vn

MO

/I 1\ 1/2

99

|3c y\ 3.643o-/V or 2.576<r I H ]

M2

Examples

Let us suppose that we have two rocks which have been selected as meeting
the specifications that qualify them for radioactive age dating. If a single
determination is to be made on each, then, if we are to have an 80% confidence
in our ability to distinguish between them at the 0.05 significance level, the
difference will have to exceed 3.9630"; apparently this is somewhere between
43% and 66% for lead-alpha, and between 18% and 48% for K-A ages.
These are rough figures, but apparently our best estimates of the current
resolving powers of the methods. It is certainly hoped that more reliable
measures of the resolution will soon become available, and at that time the
reader can bring up to date the conclusions presented here.

Suppose now that a single age determination has been made on each of the
two rocks, then with 95% confidence we can say that there is indeed a differ-
ence between their actual ages if the observed difference between the two
determinations exceeds 2.772<r, that is, it seems, 30-46% for lead-alpha and
12-33% for K-A ages. If the two determinations are each the mean of four
replicate 10 measurements, then the critical value is halved. It should be noted
that it is possible to detect a difference which is smaller than the resolution,

10 There is obviously a practical limit to the extent to which we should endeavor to
improve the precision of a method by increasing the number of replicate readings.

132 DONALD B. McIXTYRE

but the a priori confidence in our ability is less than 80% ; indeed \ve may
c% detect"' a difference which is actually zero, but we may make a type I error
in doing so !

The mean age of 11 specimens of tonalite from the batholith of southern
California is 114 m.y., whereas the mean of 7 specimens of granodiorite from
the same plutonic complex is 105 m.y. (Larsen, et al , 1958, Tables 6 and 7).
If we take a = 10% then <r- tl er ?2 = ICK^ -r f' 1 2 = 4.8%. Hence the
resolution is 2.80 X 4.8 = 13.5% or nearly 15 m.y. The critical value is
1.96 X 4.8 = 9.5% or nearly 10.5 m.y. Thus we conclude that if there is any
difference between the actual mean ages of the tonalites and granodiorites in
the batholith of southern California then, with 80% confidence, the difference
is less than 15 m.y.

Of all the Mesozoic batholiths of western North America which have been
dated by the lead-alpha method (Larsen, et al., 1958, Table 12) the widest
range of apparent mean ages is between the Sierra Nevada (n = 15, ? = 102)
and Idaho (n = 16, / = 108) batholiths. Taking <r = 10% u , we have
o- t = 2.5%. Hence the resolution is 3.96 X 2.5 = 9.9%, or nearly 10.5 m.y.
The critical value is 2.77 X 2.5 = 6.9% or nearly 7.5 m.y. Thus we must
conclude that if there is any difference between the mean ages of these batho-
liths, then, \\ith 80% confidence, it is less than 11 m.y.

The relative ages of the granitic rocks of Yosemite (Sierra Nevada) are be-
lieved to be known from their field relationships. Ten of these rocks have been
dated by the potassium-argon method (Evernden, Curtis, and Lipson, 1957).
The ages are given to 0.1 m.y., and it is reported that "The agreement between
field and experimental determinations of relative age of the various plutons is
remarkable. It can be seen that the average interval of time between intru-
sions is approximately two million years." (Curtis, et a/., 1958, p. 12)

Apparently replicate determinations were not made, with the exception
that three specimens of the Half Dome quartz monzonite were dated. That
this number is too small to yield a satisfactory estimate of the precision is
seen from the fact that, whereas s = 0.8%, the 99%-confidence range for
<r = 0.4-13.3%. Until a better estimate of the precision is published, then,
largely for the purpose of illustration, we will adopt <r = 10% as a tentative
estimate. The resolution would then be about 34 m.y., and the critical value
about 24 m.y. (see Fig. 2). If the resolution were to equal the range of observed
ages, 18.4 m.y., the analytical precision would have to be more than twice
as good or else at least five replicate determinations would have to be made on
each specimen. To bring the resolution down to one million years, the analyti-
cal precision would have to be about 30 times as good as it now appears to be.

11 The confidence levels quoted do not reflect the present uncertainty in the values
of<r.

PRECISION AND RESOLUTION IN GEOCHROXOMETRY 133

We conclude therefore that there is about a 709c probability that the observed
range of dates (about zhcr) on the Yosemite rocks is due to random errors in
the analytical methods; or alternatively, \\ith the customary confidence levels,
if there is any difference between the actual ages of the rocks, it is less than
34 m.y. If cr is as low as 5%, then the resolution would be 20 m.y. or almost
equal to the apparent range.

In this field, reproducibility tests are very costly; however, the results are
not useful until we have some estimate of their precision. Science is a rational
activity and authoritarianism is out of place; hence potential users of published
dates are entitled to know the nature, extent, and findings of any reproduci-
bility tests which have been made. Given certain assumptions and a knowl-
edge of precision, we can hope to detect discrepant observations; but without
better knowledge than seems to exist currently, it would be disquieting to find
circumstantial evidence suggesting that dates may be discarded on the basis
that they do not fit expectation. If a rock, which was acceptable a priori, is
rejected a posteriori because of its age, then full details regarding the precision
must be given. Quite apart from judgment of the harder problems of relevancy
and accuracy, no one should use a date until he has satisfied himself that the
resolution of the method justifies the use intended.

REFERENCES CITED

ALDRICH, L. T., DAVIS, G. L., TELTON, G. R., and WETHERILL, G. W., 1956, Radio-
active ages of minerals from the Brown Derby Mine and the Quartz Creek granite
near Gunnison, Colorado: J. Geophys. Res., vol. 61, pp. 215-232.

CARR, D. R. and KULP, J. L., 1957, Potassium-argon method of geochronometry : Geol.
Soc. Am., B., vol. 68, pp. 763-784.

CROW, E. L., DAVIS, F. A., and MAXFIELD, M. W., 1960, Statistics manual: New York,
Dover, 288 pp.

CURTIS, G. H., EVERNDEN, J. F., and LIPSON, J., 1958, Age determination of some
granitic rocks in California by the potassium-argon method: California Div. Mines,
Sp. Report 54, 16 pp.

and REYNOLDS, J. H., 1958, Notes on the potassium-argon dating of sedimentary

rocks: Geol. Soc. Am. B., vol. 69, pp. 151-160.

EVERNDEN, J. F., CURTIS, G. H., and LIPSON, J., 1957, Potassium-argon dating of igneous
rocks: Am. Assoc. Petroleum Geol., B., vol. 41, pp. 2120-2127.

PAUL, HENRY, 1954, Nuclear geology: New York, John Wiley, 414 pp.

FOLINSBEE, R. E., LIPSON, J., and REYNOLDS, J. H., 1956, Potassium-argon dating:
Geochim. et Cosmochim. Acta., vol. 10, pp. 60-68.

GENTNER, W., JENSEN, F., and MEHNERT, K. R., 1954, Zur geologischen Altersbestim-
mung von Kalifeldspat nach der Kalium-Argon-Methode: Zs. Naturforschung, Bd.
9a, 176 pp.

134 DOXALD B. McINTYRE

and KLEY, W., 1958, Argonbestimmungen an Kaliummineralien-V: Geochim.

et Cosmochim. Acta., vol. 14, pp. 98-104.
GOLDSCHMIDT, V. M., 1944, Crystal chemistry and geochemistry: Chemical Products,

vol. 7, pp. 29-34.
GOTTFRIED, D., JAFFE, H. \V., and SENFTLE, F. E., 1959, Evaluation of the lead-alpha

(Larsen) method for determining ages of igneous rocks: U. S. Geol. Survey, B.,

1097-A, 63 pp.
HURLEY, P. M. and FAIRS AIRN, H. \V., 1957, Abundance and distribution of uranium

and thorium in zircon, sphene, apatite, epidote, and monazite in granitic rocks:

Am. Geophys. Union, Tr., vol. 38, pp. 939-944.
LARSEN, E. S., JR., KEEVIL, N. B., and HARRISON, H. C., 1952, Method for determining

the age of igneous rocks using the accessory minerals: Geol. Soc. Am. B., vol. 63,

pp. 1045-1052.
, GOTTFRIED, D., JAFFE, H. W., and WARING, C. L., 1958, Lead-alpha ages of

the Mesozoic batholiths of western North America: U. S. Geol. Survey, B., 1070-B,

62pp.
LIPSON, J., 1958, Potassium-argon dating of sedimentary rocks: Geol. Soc. Am., B.,

vol. 69, pp. 137-150.
LYONS, J. B., JAFFE, H. W., GOTTFRIED, D., and WARING, C. L., 1957, Lead-alpha ages

of some New Hampshire granites: Am. J. Sci., vol. 255, no. 8, pp. 527-546.
MEHNERT, K. R., 1958, Argonbestimmungen an Kaliummineralien-VI: Geochim. et

Cosmochim. Acta, vol. 14, pp. 105-113.
ROSE, H., JR., and STERN, T., 1960, Spectrochemical determination of lead in zircon

for lead-alpha age measurements: Am. Mineralogist, vol. 45, pp. 1243-1256.
WARING, C. L. and WORTHING, H., 1953, A spectrographic method for determining

trace amounts of lead in zircon and other minerals: Am. Mineralogist, vol. 38, pp.

827-833.
WEBBER, G. R., HURLEY, P. M., and FAIRBAIRN, H. W., 1956, Relative ages of eastern

Massachusetts granites by total lead ratios in zircon: Am. J. Sci., vol. 254, pp. 574-583.

J. HOOVER MACKIN

University of Texas

Rational and Empirical

Methods of Investigation
in Geology 1

Most of us are concerned, and some of us have strong feelings, pro or con,
about what has been happening to geology in the past 25 years: greatly in-
creased use of nongeologic techniques in the solution of geologic problems,
such as dating by radioisotope methods; the tendency for what were special
fields of interest to become nearly or wholly independent disciplines, with
separate journals and jargon; and most of all, because it penetrates every field,
what may be called the swing to the quantitative.

At meetings of our societies, when the elder brethren gather together in hotel
rooms after the technical sessions, the discussion usually comes around to these
changes. There are apt to be sad postmortems for certain departments, once
powerful, which are now, owing to the retirement or flight of their older stal-
warts, largely staffed by dial twisters and number jugglers. It is stated, as a
scandalous sign of the times, that in certain departments geologic mapping is
considered to be, not research, but a routine operation something like survey-
ing from the point of view of an engineer and therefore not suitable as a basis

1 A preliminary draft of this paper was given as an address at the banquet of the
Branner Club during the meeting of the Cordilleran Section of the Geological Society
in Los Angeles, April 17, 1962. The text has benefitted in substance and form from
criticisms by the other authors of papers in this volume. I would like also to express
my gratitude to the following, who have read parts or all of the manuscript: Charles
Bell, Richard Blank, Howard Coombs, Ronald DeFord, Ken Fahnestock, Peter Flawn,
John Hack, Satish Kapoor, William Krumbein, Luna Leopold, Mark Meier, H. W.
Naismith, and Dwight Schmidt. Special thanks are due Frank Calkins, who did his
best to make the paper readable.

135

136 J. HOOVER MACKIN

for the doctoral thesis. There is almost always at least one sarcastic remark
per evening along the line of what our equation-minded youngsters think is
the function of the mirror on a Brunton compass: a comment or two on their
ignorance or disregard of the older literature: some skepticism as to whether
the author of a new monograph on the mechanism of mountain building had
ever been on a mountain, off a highway; and so on. This is partly banter, be-
cause we are aware that these are merely the usual misgivings of even" older
generation about the goings-on of every younger generation. But sometimes
there is evidence of real ill-feeling, which in part at least reflects a defensive
attitude; and there may be a few who seem to think that the clock ought to be
stopped that nothing new is good.

Though I am one of the elders, I often cross the hall to a concurrent session
of another group, our avant-garde, where there is an almost evangelical zeal
to quantify, and if this means abandoning the classical geologic methods of
inquiry, so much the better; where there are some who think of W. M. Davis
as an old duffer with a butterfly-catcher's sort of interest in scenery; where
there is likely to be, once in a while, an expression of anger for the oldsters who,
through their control of jobs, research funds, honors, and access to the journals,
seem to be bent on sabotaging all efforts to raise geology to the stature of a
science: where, in the urgency for change, it seems that nothing old is good.

This picture is not overdrawn, but it applies only to a small number: the
blacks and the whites, both sure of their ground. Most geologists are somewhere
in the gray between, and are beset with doubts. As for myself, I have sometimes
thought that the swing to the quantitative is too fast and too far, and that,
because a rather high percentage of the conclusions arrived at by certain meth-
ods of manipulating numerical data are superficial, or wrong, or even ludicrous,
these methods must be somehow at fault, and that we do well to stay with the
classical geologic methods. But at other times I have been troubled by ques-
tions: why the swing has been so long delayed in geology as compared with
physics and chemistry; and whether, with its relative dearth of quantitative
laws, geology is in fact a sort of subscience, as implied by Lord Kelvin's pro-
nouncement that what cannot be stated in numbers is not science. (For original
wording, and a thoughtful discussion, see Holton, 1952, p. 234.) Even more
disturbing is the view, among some of my friends in physics, that a concern
with cause-and-effect relations merely confuses the real issues in science; I will
return to this matter later. If only because of the accomplishments of the scien-
tists who hold these views, we must wonder whether our accustomed ways of
thinking are outmoded, and whether we should not drastically change our
habits of thought, or else turn in our compasses and hammers and fade away
quietly to some haven reserved for elderly naturalists.

Preparation for a talk on quantitative methods in geomorphology, as a
visiting lecturer at the University of Texas last year, forced me to examine

METHODS OF INVESTIGATION IN GEOLOGY 137

these conflicting appraisals of where we stand. 2 I suggest that two changes,
quite different but closely interlocking, are occurring at the same time and have
become confused in our thinking.

One of these changes includes an increase in the rate of infusion of new ideas
and techniques from the other sciences and from engineering, an increase in
precision and completeness of quantitative description of geologic features and
processes of all kinds, and an increased use of statistics and mechanical methods
of analyzing data. This change fits readily within the framework of the classical
geologic method of investigation, the most characteristic feature of which is
dependence on reasoning at every step; "Quantitative Zoology," by Simpson,
Roe, and Lewontin (1960) shows the way. In so far as it merely involves
doing more completely, or with more refinement, what we have always been
doing, it is evolutionary; and it is axiomatic that it is good. Some of us may
find it hard to keep abreast of new developments, but few oppose them even
privately, and even the most reactionary cannot drag his feet in public without
discredit to himself.

The other change is the introduction, or greatly increased use, of an alto-
gether different method of problem-solving that is essentially empirical. In
its purest form this method depends very little on reasoning; its most character-
istic feature, when it functions as an independent method, is that it replaces the
reasoning process by operations that are largely mechanical. Because in this
respect and others it is foreign to our accustomed habits of thought, we are
inclined to distrust it. By "we 53 I mean, of course, the conservatives of my
generation.

At least a part of the confusion in our thinking comes from a failure to dis-
tinguish between the evolutionary quantification, which is good, and the
mechanical kind of quantification, which I think is bad when it takes the place
of reasoning. It is not easy to draw a line between them because the empirical
procedures may stand alone, or they may function effectively and usefully as
parts of the classical geologic method; that is, they may replace, or be combined
in all proportions with, the reasoning processes that are the earmarks of that
method. When this distinction is recognized it becomes evident that the real
issue is not qualitative versus quantitative. It is, rather, rationality versus blind
empiricism.

2 1 was only dimly aware, until some library browsing in connection with method-
ology in the other sciences, of the extent of the scholarly literature dealing with the
history and philosophy of science. And I was surprised, as was Claude Albritton (1961),
to find that with a few noteworthy exceptions (for example, Conant, 1951, p. 269-295)
geology is scarcely mentioned in that literature. I should like to make it plain at the
outset that I am not a scholar I have only sampled a few anthologies of the history
of science. I should emphasize also that I do not presume to speak for geology; what
I say expresses the viewpoint of a single field geologist.

138 J. HOOVER MACKIN

Although the timing has been influenced by such leaders as Chayes, Hubbert,
Leopold, Krumbein, and Strahler, we are now in the swing to the quantitative
because of the explosive increase in the availability of numerical data in the
last few decades (Krumbein, 1960, p. 341), and because basic descriptive
spadework has now advanced far enough in many fields of geology to permit
at least preliminary formulation of significant quantitative generalizations.
The quantification of geology will proceed at a rapidly accelerating rate no
matter what we do as individuals, but I think the rate might be quickened a
little, and to good purpose, if the differences between the two groups on oppo-
site sides of the hall, at least those differences that arise from misunderstanding,
could be reduced. An analysis of certain quantitative methods of investigation
that are largely empirical will, I hope, serve to bring out both their merits and
limitations, and may convince some of our oldsters that although disregard
of the limitations may produce questionable results, it does not follow that
there is anything wrong with quantification, as such, nor with blind empiricism,
as such. But this is not very important tune will take care of the oldsters, soon
enough. This essay is for the youngsters the graduate students and its
purpose is to show that as they quantify, which they are bound to do, it is
neither necessary nor wise to cut loose from the classical geologic method.
Its message is the not very novel proposition that there is much good both hi
the old and the new approaches to problem-solving. A brief statement of
what I am calling the rational method will point up the contrast between it
and the empirical method, with which we are principally concerned.

The Rational Method

I'm sure that most American geologists are acquainted with our three out-
standing papers on method: G. K. Gilbert's "Inculcation of the Scientific
Method by Example," published in 1886; T. C. Chamberlin's "Method of
Multiple Working Hypotheses," published in 1897; and Douglas Johnson's
"Role of Analysis in Scientific Investigation," published in 1933. I do not
need to describe the so-called scientific method here; for present purposes I
need only remind you that it involves an interplay of observation and reason-
ing, in which the first observations suggest one or more explanations, the
working hypotheses, analysis of which leads to further observation or experi-
mentation. This in turn permits a discarding of some of the early hypotheses
and a refinement of others, analysis of which permits a discarding of data now
seen to be irrelevant to the issue, and a narrowing and sharpening of the focus
in the search for additional data that are hidden or otherwise hard to obtain
but which are of special diagnostic value; and so on and on. These steps are
spelled out in formal terms hi the papers just mentioned, and it was useful to
do that, but those who use the method all the time never follow the steps in

METHODS OF INVESTIGATION IN GEOLOGY 139

the order stated; the method has become a habit of thought that checks reason-
ing against other lines of reasoning, evidence against other kinds of evidence,
reasoning against evidence, and evidence against reasoning, thus testing both
the evidence and the reasoning for relevancy and accuracy at every stage of
the inquiry.

It now seems to be the vogue to pooh-pooh this method, as differing in no
essential way from the method of problem-solving used by the man in the street.
I've been interested in watching the way in which men in the street, including
some medical doctors practitioners, not investigators arrive at conclusions,
and I can only suggest that the scientists who insist that all persons arrive at
conclusions in the same way should reexamine their conviction. There are, of
course, rare intellects that need no disciplining, but for most of us with ordinary
minds, facility in the operations that I have just outlined must be acquired by
precept, example, and practice.

The objective of the scientific method is to understand the system investi-
gated to understand it as completely as possible. To most geologists this
means understanding of cause and effect relations within the system (Garrels,
1951, p. 32). Depending on the nature of the problem and its complexity,
quantitative data and mathematical manipulations may enter the investigation
early or late. In general, the larger the problem, the more many-sided it is,
the more complicated by secondary and tertiary feedback couples, and the more
difficult it is to obtain the evidence, the more essential it is to the efficient
prosecution of the study that the system first be understood in qualitative terms;
only this can make it possible to design the most significant experiments, or
otherwise to direct the search for the critical data, on which to base an eventual
understanding in quantitative terms.

A problem any problem when first recognized, is likely to be poorly
defined. Because it is impossible to seek intelligently for explanations until we
know what needs explaining, the first step in the operation of the scientific
method is to bring the problem into focus. This is usually accomplished by
reasoning, i.e., by thinking it through, although we will see shortly that there is
another way. Then, if it is evident that the problem is many-sided, the investi-
gator does not blast away at all sides at once with a shotgun; he shoots at one
side at a time with a rifle with the rifle, and the bullet, that he considers best
suited to that side.

This means that the investigator admits to his graphs, so to speak, only items
of evidence that are relevant to the particular matter under investigation, and
that are as accurate as practicable, with the probable limits of sampling and
experimental error expressed graphically. In reading answers from the graph,
he does no averaging beyond that required to take those limits into account.
And once an item of information has been admitted to the graph, it cannot
be disregarded; as a rule, the items that lie outside the clusters of points are at

140 J. HOOVER MAGKIN

least as significant, and usually much more interesting, than those that lie
within the clusters. It is from inquiry as to wh\ these strays are where they are
that most new ideas most breakthroughs in science develop.

The scientific method tries to visualize whole answers complete theoretical
structures at the verv outset; these are the working hypotheses that give
direction to the seeking-out and testing of evidence. But one never rushes
ahead of the data-testing process to a generalization that is regarded as a
conclusion. This is not because there is anything ethically wrong with quick
generalizing. It is only that, over a period of 500 years, investigators have
found that theoretical structures made in part of untested and ill-matched
building blocks are apt to topple sooner or later, and that piling them up and
building on them is therefore not an efficient way to make progress. The need
to test the soundness of each building block before it gets into the structure
to determine the quality* and the relevance of each item of evidence before it
gets onto the graph is emphasized by Douglas Johnson (1933). His approach
was the antithesis of that to which we mav now turn.

The Empirical Method

What I have long thought of as the engineering method or the technologic
method (we shall soon see that it needs another name) deals almost exclusively
with quantitative data from the outset, and proceeds directly to a quantitative
answer, which terminates the investigation. This method reduces to a mini-
mum, or eliminates altogether, the byplay of inductive and deductive reason-
ing by which data and ideas are processed in the scientific method; this means
that it cannot be critical of the data as they are gathered. The data are analyzed
primarily by mathematical methods, which make no distinction between cause
and effect; understanding of cause and effect relations may be interesting, but
it is not essential, and if explanations are considered at all, there is usually
only one, and it is likely to be superficial. All of the reasoning operations that
characterize the so-called scientific method depend on a fund of knowledge,
and on judgment based on experience; other things being equal, the old hand
is far better at these operations than the novice. But the operations of the
"engineering method" are much less dependent on judgment; in applying this
method the sharp youngster may be quicker and better than the experienced
oldster. For this reason and because of its quick, positive, quantitative answers,
it makes a strong appeal to the younger generation. I would like now to explain
the logic of this method, as it operates in engineering.

Many engineers feel that unless a relation can be stated in numbers, it is
not worth thinking about at all. The good and sufficient reason for this attitude
is that the engineer is primarily a doer he designs structures of various types,
and supervises their building. In the contract drawings for a bridge he must

METHODS OF INVESTIGATION IN GEOLOGY 141

specify the dimensions and strength of each structural member. Xonengineers
may be able to think of a drawing that indicates the need for a rather strong
beam at a given place in the bridge. But a young man who has spent five years
in an engineering school is incapable of thinking seriously of a "rather strong"
beam; all of the beams of his mind's eye have numerical properties. If the
strength of a beam cannot be put in numerical terms, thinking about it is mere
daydreaming.

The matter of stresses in a steel structure is fairly cut and dried. But the
engineer is confronted with many problems for which there are no ready
answers; he must deal with them he must complete his working drawings
against a deadline. If he is charged with the task of designing a canal to carry
a certain flow of irrigation water without either silting or erosion of the bed, or
with the immensely more complex task of developing and maintaining a 10-foot
navigable channel in a large river, he cannot wait until he or others have
developed a complete theory of silting and scouring in canals and rivers.
It may be 50 or a 100 years before anything approaching a complete theory,
in quantitative terms, can be formulated; and his drawings, which must be
entirely quantitative, have to be ready within a few days or weeks for the con-
tractors who will bid the job. So he has to make certain simplifying assump-
tions, even though he realizes that they may be wide of the mark, and he has
to make-do with data that are readily available, even though they are not
entirely satisfactory, or with data that can be obtained quickly from experi-
ments or models, even though the conditions are significantly different from
those existing in his particular canal or river.

He is accustomed to these expedient operations, and he is not much con-
cerned if, in plotting the data, he mixed a few oranges with the apples. In fact,
he wouldn't worry much if a few apple crates and a few orange trees got onto his
graph. He cannot scrutinize each item of evidence as to quality and relevancy;
if he did, none but the simplest of structures would ever get built. He feels
that if there are enough points on a scatter diagram, the bad ones will average
out, and that the equation for the curve drawn through the clustered points
will be good enough for use in design, always with a goodly factor of safety as
a cushion. And it almost always is. This method is quantitative^ empirical^ and
expedient. As used by the engineer, it is logical and successful.

It is of course used by investigators in many fields other than engineer-
ing. Friends in physics and chemistry tell me that- it accounts for a large per-
centage of the current research in those sciences. A recent paper by Paul Weiss
(1962) with the subtitle "Does Blind Probing Threaten to Displace Experience
in Biological Experimentation?" calls attention to its increasing use in biology.
The approach and examples are different, but the basic views of Dr. Weiss
correspond so closely with those expressed in this essay that I am inclined to
quote, not a passage or two, but the whole paper. Because this is impracticable,

142 J. HOOVER MACKIN

I can only urge that geologists interested in this phase of the general problem
whither are we drifting, methodologically? read it in the original.

In view of its widespread use in science, what I have been calling the engineer-
ing or technologic method certainly should not be identified, by name, with
engineering or technology as such. And on the other side of the coin, the
so-called scientific method is used more consistently and effectively by many
engineers and technologists than by most scientists. Besides being inappropriate
on this score, both terms have derogatory or laudatory connotations which
beg some questions. So, with serious misgivings that will be left unsaid, I will
from here on use the term "rational method" for what we are accustomed to
think of as the scientific method, and what I have been calling the engineering
method will be referred to as the empirical method. 3

Actually, the method that I am trying to describe is an empirical method;
it is shotgun or scatter-diagram empiricism, very different from the one-at-a-
time, cut-and-try empiricism of Ehrlich who, without any reasoned plan, tried
in turn 606 chemical substances as specifics for syphilis. The 606th worked.
Both the scatter-diagram and the one-at-a-time types can be, at one extreme,
purely empirical, or, if you prefer, low-level empirical. As Conant (1952,
pp. 26-30) points out, the level is raised the empirical approaches the rational
as the gathering and processing of the data are more and more controlled
by reasoning.

Use of Examples

The expositions of the rational method by Gilbert, Chamberlin, and Johnson
all depend on the use of examples, and having tried several other ways, I am
sure that this is the only way to make clear the workings of the empirical
method. I have chosen to use actual examples, because these are far more
effective than anything I could invent. They could have been selected from
any field in geology. My examples are from recent publications dealing
with the geologic work of rivers; I know of no other field hi which the two
approaches to problem-solving stand in such sharp contrast. "Horrible ex-
amples" are available, analysis of which would have a certain entertainment
value; I shall draw my examples from publications that rank as important
contributions. The principal example is from a paper that is unquestionably
the outstanding report in this field, "The hydraulic geometry of stream chan-

3 So many friends have objected to these terms that I should say that I am fully
aware that they are unsatisfactory, chiefly because they have different connotations in
different fields of study. I use them in their plain English meaning. They seem to me
to be less objectionable than any other terms, but I will not take issue with those who
think otherwise.

METHODS OF INVESTIGATION IN GEOLOGY 143

nels," by Luna Leopold and Thomas Maddock (1953). I have discussed the
methodology of geologic investigation with Leopold on numerous occasions,
and we have, in effect, agreed to disagree on some points. 4

Examples are essential in a discussion of methods, but it is difficult to work
with them. The problems of fluvial hydraulics are so complex that if the ex-
amples are to be comprehensible they must be simplified, and we must treat
them out of context. This may irritate the few who are familiar with these
matters at the technical level; I can only ask their indulgence on the ground
that I am steering a difficult course between nonessential complexity and over-
simplification. I should acknowledge, moreover, that I am an interested party;
about 15 years ago I published an article in this field (Mackin, 1948). Finally,
and most important, I will be deliberately looking at the way data are handled
from the point of view of the conservative geologist, unaccustomed to this
manner of handling data and highly critical of it. But I will come around full
circle in the end, to indicate that the operations I have been criticizing are
those of a valid method of investigation which is here to stay.

Downstream Change in Velocity in Rivers

All of us have seen the white water of a rushing mountain stream and the
smooth-surfaced flow of the streams of the plains, and we are prepared by the
contrast to suppose that the velocity of the flow decreases downstream. We are
aware, moreover, that slope commonly decreases downstream and that velocity
tends to vary directly with slope. Finally, we know from observation that the
grain size of the load carried by rivers tends to decrease downstream, and
that the grain size of the material carried by a river varies directly with some
aspect of the velocity. For these reasons, we have always taken it for granted
that velocity decreases downstream.

So in 1953, when Leopold and Maddock stated that velocity in rivers in-
creases downstream, the statement came as a first-rate shock to most geologists.
Three graphs (Fig. 1) from that article are good examples of the sort of evi-
dence, and the manner of handling evidence, on which this generalization is
based. They are log-log plots of several parameters; at the top, width of channel
against discharge in cubic feet per second; in the middle, depth against dis-
charge; and at the bottom, velocity against discharge. Each point represents
data obtained from a U. S. Geological Survey gaging station in the Yellowstone-
Big Horn drainage system. The points at the far left, such as 13 and 16, are
on small headwater tributaries, and those at the far right, such as 19, are on
the main stem of the Yellowstone. The upper and middle graphs show that,
as should be expected, both width and depth increase with increase in dis-

4 Leopold states his position elsewhere in this volume.

144

J. HOOVER MACKIX

500

100 1000

Mean annual discharge, cf$

10,000

Fie. 1. Width, depth, and velocity in relation to discharge, Bighorn and Yellowstone
Rivers, Wyoming and Montana (Leopold and Maddock, 1953, Fig. 6).

charge; the line in the lower graph also slopes up to the right; that is, velocity
increases with increase in discharge.

Some may wonder why we have moved over from increase in velocity down-
stream, which is the exciting issue, to increase in velocity with increase in dis-
charge. While it is true that discharge increases downstream in most rivers,
it is at best only an approximate measure of distance downstream the distance
that would be traveled, for example, by the grains composing the load. The
answer given in the Leopold-Maddock paper is that there were not enough
gaging stations along the rivers to provide a sufficient number of points. Use
of discharge, rather than distance, makes it possible to bring onto one graph
the main stream and its tributaries of all sizes; or, for that matter, since "main

METHODS OF INVESTIGATION IN GEOLOGY 145

stream" is a relative term, all the neighboring streams in an area large enough
to provide enough points to bring out the significant relationships.

This explanation does not quite answer the question, unless expediency is an
answer, but it raises another question.

Velocity at any given place at any gaging station, for example varies with
variations in discharge from time to time during the year; as discharge and
depth increase, usually in the spring, velocity at a given place increases very
markedly. We may ask, then, what discharge is represented by the points on
the lower graph? The question is pertinent, because we know that in most
rivers much of the year's transportation of bed load the sand and gravel that
move along the bed is accomplished during a relatively brief period of maxi-
mum discharge. But these graphs show mean annual discharges, and the
velocities developed at those discharges. The reason for using mean annual
discharge is said to be that this parameter is readily available at a large number
of gaging stations. This explanation does not answer the question: what is the
relevance of mean annual discharge in an analysis of the geologic work of
rivers?

This general question, which applies to each of the stations considered indi-
vidually, takes on another meaning when the relations between mean annual
discharge and maximum discharge on streams are considered. Reference to
Water Supply Paper 1559 (1960, p. 169) indicates that at point 13 (Fig. 1),
which represents a gaging station on the North Fork of Owl Creek, the average
annual discharge for the 14-year period of record was 15 cfs (cubic feet per
second), whereas the maximum discharge during the same period was 3200 cfs;
that is, the maximum was about 213 times the average. The same paper
(p. 234) indicates that at point 19, which represents the Yellowstone River at
Sidney, Montana, the average annual discharge over a 46-year period was
13, 040 cfs, whereas the maximum during the same period was 15 9, 000 cfs;
here the maximum was about 12 times the average. The noteworthy thing
about this graph the thing that makes it so exciting is that it shows that
velocity increases downstream although we know from observation that grain
size decreases downstream. The significance of the graph is more readily under-
stood when we remember: (1) that the larger grains move only at times of
maximum discharge; (2) that this graph shows mean annual discharge; and (3)
that in the small rivers on the left side, the maximum discharge may be more
than 200 times as great as the discharge shown on the graph, while in the big
rivers on the right, it is less than 20, and usually less than 10 times the discharge
shown, that is, that the critical ratios on the two sides are of a different order of
magnitude. The slope of the line is an important statistical fact, but it does
not bear directly on transportation of bed load by rivers.

One more thought in this connection. The depth at average annual dis-
charge at point 13, on the North Fork of Owl Creek, is shown in the middle

146 J. HOOVER MACKIN

graph as being something less than 0.6 foot. I know the general area, and,
although I have no measurements at this gaging station, it is my recollection
that the larger boulders on the bed of the North Fork are more than 0.6 foot
in diameter the boulders on the bed have diameters that are of the same order
of magnitude as the depth at which the very low velocity shown for this point
was calculated. Similar relationships obtain for other small headwater streams,
the points for which anchor down, so to speak, the left end of the line.

Let us look briefly at one more aspect of the case. The velocity is lower near
the bed of a river than near the surface. Rubey (1938) and others have shown
that the movement of bed load is determined, not by the average velocity, but
by the velocity near the bed. And it has also been established that the relation
between average velocity and what Rubey calls "bed velocity" varies markedly
with depth of water, roughness of channel, and other factors. We may reason-
ably ask, then, what velocity is represented by the points on the graph? The
answer is spelled out clearly by Leopold and Maddock (1953, p. 5).

Velocity discussed in this report is the quotient of discharge divided by
the area of the cross section, and is the mean velocity of the cross section as
used in hydraulic practice . . . This mean velocity is not the most meaningful
velocity parameter for discussing sediment transport, but it is the only
measure of velocity for which a large volume of data is available. Although
the writers recognize its limitations, the mean velocity is used here in lieu of
adequate data on a more meaningful parameter.

There are various other similar questions about this graph, some of which
are discussed by the authors in the clear and candid style of the last quotation.
I will not develop these questions, or the secondary and tertiary questions that
spring from the answers. Some of you may be thinking: never mind the indi-
vidual points; what about the trends? It could be argued that if the conclusions
are internally consistent; if they match those for other river systems; if, in
short, these procedures get results, this alone justifies them.

Let's look at the results. Figure 2 is the velocity-discharge graph of Fig. 1,
modified by use of symbols to identify related points and with dashed lines for
individual rivers.

Points 1, 2, 3, 4, and 5, are on the main stem of the Big Horn River. Points
1, 2, and 3 are in the Big Horn Basin; point 4 is about 50 miles downstream
from 3, and 5 is about 20 miles downstream from 4. The dashed line, which
fits these points quite well, slopes down to the right; it means that on the main
stem of the Big Horn, velocity decreases downstream.

Points 6, 7, 8, and 9 are on the Wind River, which is actually the upper
part of the Big Horn River. I do not know whether points 8 and 9 represent
the same types of channel conditions as 1, 2, 3, 4, and 5, as suggested by their
positions, or whether they should be grouped with 6 and 7, as called for by the

METHODS OF INVESTIGATION IN GEOLOGY

147

10

100 1000

Mean annual discharge, cfs

10,000

FIG. 2. Same as velocity-discharge graph in Fig. 1, with dashed lines for certain
rivers; triangles, Greybull River; open circles, Wind River; X's, Bighorn River; solid
squares, Yellowstone River.

geographic usage of the names. Let us say, then, that in what the geographers
call the Wind River, velocity at average annual discharge increases downstream.

Points 10 and 11, both on the Greybull River, also suggest by their relative
position that velocity increases downstream; the line slopes up to the right.
But point 10 is near the mouth of the Greybull, about 40 airline miles down-
stream from 9; the average annual discharge decreases downstream (Leopold,
1953, p. 612) partly because of withdrawal of water for irrigation. Velocity
actually decreases very markedly downstream on the Greybull.

Points 17, 18, and 19 are on the main stem of the Yellowstone. It appears
from this graph that velocity increases downstream between 17 and 18, and
decreases downstream between 18 and 19.

The generalization that velocity increases downstream, at a rate expressed
by the slope of the solid line on this graph, is a particular type of empirical
answer. It is what the nonstatistician is likely to think of as an "insurance
company 35 type of statistic a generalization applying to this group of rivers
collectively, but not necessarily to any member of the group. Of the river
segments represented on the graph, about half increase in velocity down-
stream, and about half decrease in velocity downstream. As shown by the
different slopes of the dashed lines, in no two of them is the rate of change in
velocity the same.

This is really not very surprising. The solid line averages velocity-discharge
relations in river segments that are, as we have seen, basically unlike in this
respect. Moreover, slope, which certainly enters into velocity, is not on the
graph at all. For these reasons the equation of the solid line is not a definitive
answer to any geologic question.

But and here I change my tune this graph was not intended to provide
a firm answer to any question. It is only one step a preliminary descriptive
step in an inquiry into velocity changes in rivers from head to mouth. This
is accomplished by plotting certain conveniently available data on a scatter
diagram.

148 J. HOOVER MACKIX

I have indicated earlier how this procedure, which is empirical, expedient,
and quantitative, serves the practicing engineer very well in getting answers
that are of the right order of magnitude for use in design in deadline situations.
Here we see the same procedure operating as a step in a scientific investigation.
It is used in this graph to learn something about velocity relations in rivers
from a mass of data that were obtained for a different purpose; the purpose
of U. S. Geological Survey gaging stations is to measure discharge, not velocity.
This gleaning of one kind of information from measurements particularly
long-term records of measurements that are more or less inadequate because
they were not planned to provide that kind of information, is a very common
operation in many scientific investigations, and is altogether admirable.

There is another point to be made about this graph. Before the work repre-
sented by it was done, there had been no comprehensive investigations of
velocity in rivers from head to mouth: this study was on the frontier. In these
circumstances, some shots in the dark some shotgun shots in the dark were
quite in order. The brevity with which this point can be stated is not a measure
of its importance.

Finally, I wish to emphasize that Leopold and Maddock did not regard the
solid line as an answer its equation was not the goal of their investigation.
They went on in this same paper, and in others that have followed it, to deal
with velocities developed at peak discharges and with many other aspects of
the hydraulic geometry of river channels. It is for this reason, and the other
two reasons just stated, that I can use the graph as I have without harm to its
authors.

But our literature is now being flooded by data and graphs such as these,
without any of the justifications, engineering or scientific, that I have outlined.
In many instances the graph is simply a painless way of getting a quantitative
answer from a hodge-podge of data, obtained in the course of the investigation,
perhaps at great expense, but a hodge-podge nevertheless because of the failure
of the investigator to think the problem through prior to and throughout the
period of data gathering. The equations read from the graphs or arrived at
by other mechanical manipulations of the data are presented as terminal scien-
tific conclusions. I suggest that the equations may be terminal engineering
conclusions, but, from the point of view of science, they are statements of prob-
lems, not conclusions. A statement of a problem may be very valuable, but if
it is mistaken for a conclusion, it is worse than useless because it implies that
the study is finished when in fact it is only begun.

If this empirical approach this blind probing were the only way of quan-
tifying geology, we would have to be content with it. But it is not; the quanti-
tative approach is associated with the empirical approach, but it is not wedded
to it. If you will list mentally the best papers in your own field, you will dis-
cover that most of them are quantitative and rational. In the study of rivers

METHODS OF INVESTIGATION IN GEOLOGY 149

I think of Gilbert's field and laboratory studies of Sierra Nevada mining debris
(1914, 1917), and Rubey's analysis of the force required to move particles on
a stream bed (1938). These geologists, and many others that come to mind,
have (or had) the happy faculty of dealing \\ith numbers \vithout being carried
away by them of quantifying without, in the same measure, taking leave
of their senses. I am not at all sure that the percentage of geologists capable
of doing this has increased very much since Gilbert's day. I suggest that
an increase in this percentage, or an increase in the rate of increase, is in the
direction of true progress.

We shall be seeing more and more of shotgun empiricism in geologic writings,
and perhaps we shall be using it in our own investigations and reports. We must
learn to recognize it when we see it, and to be aware of both its usefulness and
its limitations. Certainly there is nothing wrong with it as a tool, but, like most
tools, how well it works depends on how intelligently it is used.

Causes of Slope of the Longitudinal Profile

We can now turn to a matter which seems to me the crux of the difference
between the empirical and the rational methods of investigation, namely,
cause-and-effect relations. 5 I would like to bring out, first, an important
difference between immediate and superficial causes as opposed to long-term,
geologic causes; and second, the usefulness, almost the necessity, of thinking
a process through, back to the long-term causes, as a check on quantitative
observations and conclusions.

Most engineers would regard an equation stating that the size of the pebbles
that can be carried by a river is a certain power of its bed velocity as a com-
plete statement of the relationship. The equation says nothing about cause

5 I am aware that my tendency to think in terms of cause and effect would be regarded
as a mark of scientific naivete by some scientists and most philosophers. My persistence
in this habit of thought after having been warned against it does not mean that I chal-
lenge their wisdom. Perhaps part of the difficulty lies in a difference between what I
call long-term geologic causes and what are sometimes called ultimate causes. For
example, a philosopher might say, "Yes, it is clear that such things as discharge and size
of pebbles may control or cause the slope of an adjusted river, but what, then, is the
cause of the discharge and the pebble size? And if these are effects of the height of the
mountains at the headwaters, what, then, is the cause of the height of the mountains"?
Every cause is an effect, and every effect is a cause. Where do we stop? I can only
answer that I am at the moment concerned with the geologic work of rivers, not with
the cause of upheaval of mountains. The question, where do we stop?, is for the phi-
losopher, who deals with all knowledge; the quest for ultimate causes, or the futility
of that quest, is in his province. The investigator in science commonly stays within
his own rather narrow field of competence and, especially if time is an important
element of his systems, he commonly finds it useful to think in terms of cause and effect
in that field. The investigator is never concerned with ultimate causes.

150 J. HOOVER MACKIN

and effect, and the engineer might be surprised if asked which of the two,
velocity or grain size, is the cause and which is the effect. He would almost
certainly reply that velocity controls or determines the size of the grains that
can be moved, and that therefore velocity is the cause. To clinch this argu-
ment, he might point out that if, by the turn of a valve, the velocity of a labora-
tory river were sufficiently increased, grains that previously had been at rest
on the bed would begin to move; that is, on the basis of direct observation, and
by the commonsense test of relative timing, the increase in velocity is the cause
of the movement of the larger grains. This is as far as the engineer needs to
go in most of his operations on rivers.

He might be quite willing to take the next step and agree that the velocity
is, in turn, partly determined by the slope. In fact, getting into the swing of
the cause-and-effect game, he might even volunteer this idea, which is in terri-
tory familiar to him. But the next question what then, is the cause of the
slope? leads into unfamiliar territory; many engineers, and some geologists,
simply take slope for granted.

Our engineer would probably be at first inclined to question the sanity of
anvone suggesting that the size of the grains carried by a river determines the
velocity of the river. But in any long-term view, the sizes of the grains that are
supplied to a river are determined, not by the river, but by the characteristics
of the rocks, relief, vegetative cover, and other physical properties of its drain-
age basin. If the river is, as we say, graded (or as the engineer says, adjusted),
this means that in each segment the slope is adjusted to provide just the trans-
porting power required to carry through that segment all the grains, of what-
ever size, that enter it from above. Rivers that flow from rugged ranges of
hard rock tend to develop steep slopes, adjusted to the transportation of large
pebbles. Once they are developed, the adjusted slopes are maintained indefi-
nitely, as long as the size of the pebbles and other controlling factors remain the
same. Rivers that are supplied only with sand tend to maintain low slopes
appropriate to the transportation of this material.

If the sizes of the grains supplied to a given segment of an adjusted river are
abruptly increased by uplift, by a climatic change, or by a work of man, the
larger grains, which are beyond the former carrying power, are deposited in
the upper part of the segment; the bed is raised thereby and the slope is conse-
quently steepened. This steepening by deposition continues until that particular
slope is attained which provides just the velocity required to carry those larger
grains, that is, until a new equilibrium slope is developed, which the river will
maintain thence forward so long as grain size and other slope-controlling
conditions remain the same.

Thus in the long view, velocity is adjusted to, or determined by, grain size;
the test of relative timing (first the increase in grain size of material supplied
to the river, and then, through a long period of readjustment, the increase in

METHODS OF INVESTIGATION IN GEOLOGY 151

velocity) marks the change in grain size as the cause of the change in velocity.
Note that because the period of readjustment may occupy thousands of years,
this view is based primarily on reasoning rather than on direct observation.
Note also that we deal here with three different frames of reference spanning
the range from the empirical to the rational.

The statement that grain size tends to vary directly with bed velocity is an
equation, whose terms are transposable; neither time nor cause and effect are
involved, and this first frame may be entirely empirical. The numerical
answer is complete in itself.

The short-term cause-and-effect view, that grain size is controlled by bed
velocity, is in part rational, or if you prefer, it represents a higher level of
empiricism. As I see it, this second frame has a significant advantage over the
first in that it provides more fertile ground for the formulation of working
hypotheses as to the mechanical relations between the flow and the particle at
rest or in motion on the bed, leading to purposeful observation or to the design
of experiments.

The third frame, the long-term view, that velocity is controlled by grain
size, has a great advantage over the short-term view in that it provides an un-
derstanding of the origin of slope, which the short-term view does not attempt
to explain. It is largely rational, or if you prefer, it represents a still higher level
of empiricism.

Because I think that the objective of science is an understanding of the world
around us, I prefer the second and third frames to the first, but I hope that it is
clear that I recognize that all the frames are valid; the best one, in every
instance, is simply the one that most efficiently gets the job done that needs
doing. The important thing is to recognize that there ore different frames;
and that they overlap so completely and are so devoid of boundaries that it is
easy to slip from one to the other.

The difference between the rational and the empirical approach to this
matter of river slope, and the need for knowing what frame of reference we are
in, can be clarified by a little story. One of the earliest theories of the origin
of meanders, published in a British engineering journal in the late eighteen
hundreds, was essentially as follows: divested of all geographic detail. Two
cities A and B, both on the valley floor of a meandering river, are 50 airline
miles apart. City B is 100 feet lower than city A\ hence the average slope of
the valley floor is two feet per mile (Fig. 3). But the slope of the river, measured
round its loops, is only one foot per mile. The British engineer's theory was,
in effect, though not expressed in these words, that the river said to itself,
"How, with a slope of one foot per mile, can I manage to stay on a valley floor
with a slope of two feet per mile? If I flow straight down the middle of the
valley floor, starting at A, I will be 50 feet above the valley floor at 2?, and that
simply will not do." Then it occurred to the river that it could meet this

152 J. HOOVER MACKIN

FIG. 3. Diagram illustrating an hypothesis for the origin of meanders.

problem by bending its channel into loops of precisely the sinuousity required
to keep it on the valley floor, just as a man might do with a rope too long for
the distance between two posts. And it worked, and that's why we have
meanders.

Note that this theory not only explains meandering qualitatively, but puts
all degrees of meandering, from the very loopy meanders of the ribbon-candy
type to those that are nearly straight, on a firm quantitative basis the sinu-
osity or degree of meandering, M, equals the slope of the valley floor, S y ,
over the slope of the river, S T .

There is nothing wrong with this equation, so long as it only describes.
But if its author takes it to be an explanation, as the British engineer did, and
if he slips over from the empirical frame into the rational frame, as he may do
almost without realizing it, he is likely to be not just off by an order of magni-
tude, but upside-down to be not only wrong but ludicrous. This explana-
tion of meanders leaves one item out of account the origin of the valley floor.
The valley floor was not opened out and given its slope by a bulldozer, nor is
it a result of special creation prior to the creation of the river. The valley
floor was formed bv the river that flows on it.

Causes of Downvalley Decrease in Pebble Size

It is a matter of observation that there is commonly a downvalley decrease
in the slopes of graded rivers, and it is also a matter of observation that there
is commonly a downvalley decrease in the size of pebbles in alluvial deposits.
A question arises, then, as to whether the decrease in slope is caused in part
by the decrease in pebble size, or whether the decrease in pebble size is caused
in part by the decrease in slope, or whether both of these changes are independ-
ent or interdependent results of some other cause. My third and last example
applies the empirical and rational approaches to a part of this problem, namely,
what are the causes of the decrease in pebble size? The reasoning is somewhat
more involved than in the other examples; in this respect it is more truly rep-
resentative of the typical geologic problem.

METHODS OF INVESTIGATION IX GEOLOGY 153

The downvalley decrease in pebble size could be caused by either of two
obvious, sharply contrasted mechanisms: (1) abrasional wear of the pebbles
as they move along the bed of the stream, and (2 ) selective transportation, that
is, a leaving-behind of the larger pebbles. The question is, which mechanism
causes the decrease, or, if both operate, what is their relative importance?

There is no direct and satisfactory' way of obtaining an answer to this ques-
tion by measurement, however detailed, of pebble sizes in alluvial deposits.
The most commonly used approach is by means of laboratory experiment.
Usually fragments of rock of one or more kinds are placed in a cylinder which
can be rotated on a horizontal axis and is so constructed that the fragments
slide, roll, or drop as it turns. The fragments are remeasured from time to
time to determine the reduction in size, the corresponding travel distance
being calculated from the circumference of the cylinder and the number of
rotations. This treatment does not approximate very closely the processes of
wear in an actual river bed. Kuenen (1959) has recently developed a better
apparatus, in which the fragments are moved over a concrete floor in a circular
path by a current of water. Whatever the apparatus, it is certain that the
decrease in pebble size observed in the laboratory is due wholly to abrasion,
because none of the pebbles can be left behind; there is no possibility of
selective transportation.

When the laboratory rates of reduction in pebble size per unit of travel
distance are compared with the downvalley decrease in pebble size in alluvial
deposits along most rivers, it is found that the decrease in size along the rivers
is somewhat greater than would be expected on the basis of laboratory data on
rates of abrasion. If the rates of abrasion in the laboratory correctly represent
the rates of abrasion in the river bed, it should be only necessary to subtract
to determine what percentage of the downvalley decrease in grain size in the
alluvial deposits is due to selective transportation.

Field and laboratory data bearing on this problem have been reviewed by
Scheidegger (1961) in his textbook, "Theoretical Geomorphology," which is
about as far out on the quantitative side as it is possible to get. Scheidegger
(p. 175) concludes that ". . . the most likely mechanism of pebble gradation
in rivers consists of pebbles becoming contriturated due to the action of fric-
tional forces, but being assigned their position along the stream bed by a sorting
process due to differential transportation."

If I understand it correctly, this statement means that pebbles are made
smaller by abrasion, but that the downvalley decrease in pebble size in alluvial
deposits is due largely (or wholly?) to selective transportation.

On a somewhat different basis the rate of reduction of pebbles of less re-
sistant rock, relative to quartzite, in a downvalley direction in three rivers
east of the Black Hills Plumley (1948) concludes that about 25 per cent of

154

J. HOOVER MACKIN

the reduction in these rivers is due to abrasion, and about 75 per cent is due to
selective transportation.

These two conclusions as to the cause of the downstream decrease in pebble
size, solidly based on measurements, agree in ascribing it mainly to selective
transportation. Let us try a different approach let us think through the long-
term implications of the processes.

Downstream decrease in pebble size by selective transportation requires that
the larger pebbles be left behind permanently. The three-inch pebbles, for
example, move downstream to a certain zone, and are deposited there because
they cannot be transported farther. The two-inch pebbles are carried farther
downstream, to be deposited in an appropriate zone as the slope decreases.
These zones may have considerable length along the stream, they may be
poorly defined, and they may of course overlap, but there is a downstream
limit beyond which no pebbles of a given size occur in the alluvial deposits
because none could be carried beyond that limit, which is set by transporting
power.

Consider a river carrying a bed load of sand and gravel under steady-state
conditions such that the slope and altitude in a given segment are maintained
indefinitely without change, and let it be assumed for simplicity that the
channel is floored and walled by rock (Fig. 4a). The load moves chiefly

^TO*^

(c)

Flo. 4. Diagram illustrating exchange in graded and aggrading rivers.

METHODS OF INVESTIGATION IN GEOLOGY 155

during high-water stages and lodges on the bed during low-water stages. The
smaller pebbles are likely to be set in motion sooner than the larger pebbles
during each rising stage, they are likely to move faster while in motion, and
they are likely to be kept in motion longer during each falling stage. In this
sense, the transportation process is selective if a slug of gravel consisting of
identifiable pebbles were dumped into the segment, the smaller pebbles would
outrun the larger, and this would cause a downstream decrease in the sizes of
these particular pebbles in the low-water deposits. But in the steady-state con-
dition, that is, with a continuous supply of a particular type of pebble or of
pebbles of all types, all the pebbles deposited on the bed during the low stages
must be placed in motion during the high stages; if the larger pebbles were
permanently left behind during the seasonal cycles of deposition and erosion, the
bed would be raised, and this, in turn, would change the condition. A non-
aggrading river flowing in a channel which is floored and walled by rock cannot
rid itself of coarse material by deposition because there is no place to deposit it
where it will be out of reach of the river during subsequent fluctuations of flow;
every pebble entering a given segment must eventually pass on through it.
The smaller pebbles move more rapidly into the segment than the larger
pebbles, but they also move more rapidly out of it. In the steady-state condi-
tion, the channel deposits from place to place in the segment contain the same
proportions of the smaller and larger pebbles as though all moved at the same
rate. Selective transportation cannot be a contributing cause of a downstream
decrease in pebble size hi our model river because there can be no selective
deposition.

In a real river that maintains the same level as it meanders on a broad
valley floor, bed load deposited along the inner side of a shifting bend is ex-
changed for an equal volume of slightly older channel deposits eroded from
the outside of the bend. If these channel deposits were formed by the same
river, operating under the same conditions and at the same level over a long
period of time (Fig. 4b), the exchange process would not cause a reduction
in the gram size of the bed load; insofar as selective transportation is concerned,
the relation would be the same as hi our model river. But if, by reason of
capture or climatic change or any other change in controlling conditions, the
older alluvial deposits in a given segment are finer grained than the bed load
now entering that segment (Fig. 4c), exchange will cause a decrease in pebble
size in a downstream direction, at least until the older deposits have been
completely replaced by deposits representing the new regime. Exchange also
causes a reduction hi grain size if the river, maintaining the same level, cuts
laterally into weak country rock that yields material finer in grain size than
the load that is being concomitantly deposited on the widening valley floor.

The selective transportation associated with the process of exchange in
the graded river, while by no means negligible, is much less effective as a

156 J. HOOVER MACKIX

cause of downstream decrease in pebble size than the selective transportation
that characterizes the aggrading river. The essential difference is shown in
Fig. 4(dj: some of the deposits formed by one swing of the aggrading river
across its valley floor are not subject to reworking in later swings, because the
channel is slowly rising. The largest pebbles in transit in a given segment in a
high-\\ater stage are likely to be concentrated in the basal part of the deposit
formed during the next falling stage. Thus the aggrading river rids itself of
these pebbles, selectively and permanently, and there is a corresponding down-
stream decrease in pebble sizes in the deposits.

If upbuilding of the flood plain by deposition of overbank material keeps
pace with aggradational rising of the channel, the shifting meanders may
exchange channel deposits for older alluvium consisting wholly or in part of
relatively fine-grained overbank material (Fig. 4d). But in rapidly aggrading
rivers this rather orderly process may give way to a fill-spill mechanism in which
filling of the channel is attended by the splaying of channel deposits over ad-
joining parts of the valley floor. On some proglacial outwash plains this type
of braiding causes boulder detritus near the ice front to grade into pebbly sand
within a few miles; there is doubtless some abrasional reduction in grain size
in the proglacial rivers, but nearly all the decrease must be due to selective
transportation.

Briefly then, thinking the process through indicates that the downstream
decrease in grain size in river deposits in some cases may be almost wholly due
to abrasion, and in others almost wholly due to selective transportation, depend-
ing primarily on whether the river is graded or aggrading and on the rate of
aggradation. It follows that no generalization as to the relative importance of
abrasion versus selective transportation in rivers all rivers has any mean-
ing.

A different way of looking at this problem has been mentioned in another
connection. As already noted, selective transportation implies permanent
deposition, for example, the three-inch pebbles in a certain zone, the two-inch
pebbles in another zone farther downstream, and so on. If this deposition is
caused by a downstream decrease in slope, as is often implied and sometimes
stated explicitly (Scheidegger, p. 171), then what is the cause of the decrease
in slope? We know that the valley floor was not shaped by a bulldozer, and we
know that it was not formed by an act of special creation before the river began
to flow. As we have seen in considering the origin of meanders, rivers normally
shape their own valley floors. If the river is actively aggrading, this is usually
because of some geologically recent change such that the gradient in a given
segment is not steep enough to enable the river to move through that segment
all of the pebbles entering it; in this (aggrading) river, the size of the pebbles
that are carried is controlled in part by the slope, and the larger pebbles are
left behind. But if the river is graded, the slope in each segment is precisely

METHODS OF INVESTIGATION IN GEOLOGY 157

that required to enable the river, under the prevailing hydraulic conditions,
whatever they may be, to carry the load supplied to it. The same three-inch
pebbles that are the largest seen on the bed and banks in one zone will, after a
while, be the two-inch pebbles in a zone farther downstream.

We cannot wait long enough to verify this conclusion by direct observation
of individual pebbles, because the pebbles ordinarily remain at rest in alluvial
deposits on the valley floor for very long intervals of time between jogs of move-
ment in the channel. We are led to the conclusion by reasoning, rather than
by direct observation. In the long-term view, the graded river is a transporta-
tion system in equilibrium, which means that it maintains the same slope so
long as conditions remain the same. There is no place in this self-maintaining
system for permanent deposits: if the three-inch pebbles entering a given zone
accumulated there over a period of geologic time, they would raise the bed
and change the slope. As the pebbles, in their halting downvalley movement
in the channel, are reduced in size by abrasion, and perhaps also by weathering
while they are temporarily at rest in the valley floor alluvium, the slope, which
is being adjusted to their transportation, decreases accordingly.

Does this reasoning settle the problem? Of course not ! It merely makes us
take a more searching look at the observational data. Since it is theoretically
certain that the mechanisms which cause pebbles to decrease in size as they
travel downstream operate differently, depending on whether the river is
graded or aggrading, there is no sense in averaging measurements made along
graded rivers with those made along aggrading rivers. However meticulous
the measurements, and however refined the statistical treatment of them, the
average will have no meaning. 6

The reasoning tells us that, first of all, the rivers to be studied in connection
with change of pebble size downstream must be selected with care. Because a
steady-state condition is always easier to deal with quantitatively than a shifting
equilibrium, it would be advisable to restrict the study, at the outset, to the
deposits of graded rivers; when these are understood, we will be ready to deal
with complications introduced by varying rates of aggradation. Similarly, it
will be well, at least at the beginning, to eliminate altogether, or at least reduce
to a minimum, the complicating effects of contributions from tributaries or
other local sources; this can be done by selecting river segments without large
tributaries, or by focusing attention on one or more distinctive rock types from
known sources. There are unavoidable sampling problems, but some of these

6 I owe to Frank Calkins the thought that, like most hybrids, this one would be
sterile. The significance of this way of expressing what I have been saying about the
averaging of unlike things is brought out by Conant's (1951, p. 25) definition of science
as "an interconnected series of concepts and conceptual schemes that have developed
as a result of experimentation and observation and are fruitful of further experimenta-
tion and observations. In this definition the emphasis is on the word 'fruitful'."

158 J. HOOVER MACKIX

can readily be avoided; for example, there are many river segments in which
the alluvial deposits are not contaminated by lag materials. Any attempt to
develop sampling procedures must take into account, first of all, the fact that
the channel deposits in a given segment of a valley differ significantly in grada-
tion of grain size from the material moving through the channel in that segment
in any brief period; the investigation may deal with the bed load (trapped in a
box, so to speak), or with the deposits, or with both; but if both bed load and
the deposits are measured, the measurements can only be compared, they can-
not be averaged. Certainly we must investigate, in each river individually, the
effects of weathering of the pebbles during periods of rest.

\Ve must also take another hard look at the abrasion rales obtained by
laboratory- experiments, and try to determine in what degree these are directly
comparable with abrasion rates in rivers. It is clearly desirable to develop other
independent checks, such as those given by Plumley's measurement of rates
of downstream reduction in sizes of pebbles of rock types differing in resistance
to abrasion. Finally, it goes \\ithout saying that the reasoning itself must be
continuously checked against the evidence, and one line of reasoning must be
checked against others, to make sure that the mental wheels have not slipped
a cog or two.

\Vhen we eventually have sufficient data on rates of downstream decrease of
pebble size in alluvial deposits along many different types of rivers (considered
individually), it will be possible to evaluate separately, in quantitative terms,
the effect of special circumstances influencing the process of exchange in graded
rivers, rates of aggradation in aggrading rivers, and the other causes of down-
stream decrease in pebble size. These generalizations will apply to all river
deposits, modern as well as ancient, and it may even be that we can draw sound
inferences regarding the hydraulic characteristics of the ancient rivers by com-
paring their deposits with those of modern rivers, in which the hydraulic
characteristics can be measured.

This rational method of problem-solving is difficult and tortuous, but the
history of science makes it clear, again and again, that if the system to be in-
vestigated is complex, the longest way 'round is the shortest way home;
most of the empirical shortcuts turn out to be blind alleys.

Whither Are We Drifting, Methodologically?

I would like now to return to some of the questions asked at the outset.
Must we accept, as gospel, Lord Kelvin's pronouncement that what cannot be
stated in numbers is not science? To become respectable members of the scien-
tific community, must we drastically change our accustomed habits of thought,
abandoning the classic geologic approach to problem-solving? To the extent
that this approach is qualitative, is it necessarily loose, and therefore bad?

METHODS OF INVESTIGATION IN GEOLOGY 159

Must we now move headlong to quantify our operations on the assumption
that whatever is quantitative is necessarily rigorous and therefore good?

Why has the swing to the quantitative come so late? Is it because our early
leaders, men such as Hutton, Lyell, Agassiz, Heim, Gilbert, and Davis, were
intellectually a cut or two below their counterparts in classical physics? There
is a more reasonable explanation, which is well known to students of the history
of science. In each field of study the timing of the swing to the quantitative
and the present degree of quantification are largely determined by the subject
matter: the number and complexity of the interdependent components involved
in its systems, the relative ease or difficulty of obtaining basic data, the suscepti-
bility of those data to numerical expression, and the extent to which time is an
essential dimension. The position of geology relative to the basic sciences has
been stated with characteristic vigor by Walter Bucher (1941) in a paper that
seems to have escaped the attention of our apologists.

Classical physics was quantitative from its very beginning as a science; it
moved directly from observations made hi the laboratory under controlled
conditions to abstractions that were quantitative at the outset. The quantifi-
cation of chemistry lagged 100 years behind that of physics. The chemistry
of a candle flame is of an altogether different order of complexity from the
physics of Galileo's rolling ball; the flame is only one of many types of oxidation;
and oxidation is only one of many ways in which substances combine. There
had to be an immense accumulation of quantitative data, and many minor
discoveries some of them accidental, but most of them based on planned
investigations before it was possible to formulate such a sweeping generaliza-
tion as the law of combining weights.

If degree of quantification of its laws were a gage of maturity in a science
(which it is not), geology and biology would be 100 to 200 years behind chem-
istry. Before Bucher (1933) could formulate even a tentative set of "laws" for
deformation of the earth's crust, an enormous descriptive job had to be well
under way. Clearly, it was necessary to know what the movements of the crust
are before anybody could frame explanations of them. But adequate description
of even a single mountain range demands the best efforts of a couple of genera-
tions of geologists, with different special skills, working in the field and the
laboratory. Because no two ranges are alike, the search for the laws of mountain
growth requires that we learn as much as we can about every range we can
climb and also about those no longer here to be climbed; the ranges of the
past, which we must reconstruct as best we can by study of their eroded stumps,
are as significant as those of the present. Rates of growth and relative ages of
past and present ranges are just as important as their geometry; the student of
the mechanics of crustal deformation must think like a physicist and also like
a historian, and these are very different ways of thinking, difficult to combine.
The evidence is hard to come by, it is largely circumstantial, and there is never

160 J. HOOVER MACKIX

enough of it. Laboratory models are helpful only within narrow limits. So it
is also with the mechanism of emplacement of batholiths, and the origin of
ore-forming fluids, and the shaping of landforms of all kinds, and most other
truly geologic problems.

It is chiefly for these reasons that most geologists have been preoccupied
with manifold problems of description of geologic things and processes particu-
lar things and processes and have been traditionally disinclined to generalize
even in qualitative terms. Because most geologic evidence cannot readily be
stated in numbers, and because most geologic systems are so complex that some
qualitative grasp of the problem must precede effective quantitative study, we
are even less inclined to generalize in quantitative terms. Even-body knows
the story of Lord Kelvin's calculation of the age of the earth.

These things are familiar, but they are worth saying because they explain
why geology is only now fully in the swing to the quantitative. Perhaps it would
have been better if the swing had begun earlier, but this is by no means certain.
A meteorologist has told me that meteorology might be further ahead today if
its plunge to the quantitative had been somewhat less precipitous if there
had been a broader observational base for a qualitative understanding of its
exceedingly complex systems before these were quantified. At any rate, it is
important that we recognize that the quantification of geology is a normal
evolutionary process, which is more or less on schedule. The quantification
will proceed at an accelerating pace, however much our ultraconservatives
may drag their feet. I have been trying to point out that there is an attendant
danger: as measurements increase in complexity and refinement, and as mathe-
matical manipulations of the data become more sophisticated, these measure-
ments and manipulations may become so impressive in form that the investi-
gator tends to lose sight of their meaning and purpose. 7

This tendency is readily understandable. Some of the appealing features of
the empirical method have already been mentioned. Moreover, the very act
of making measurements, in a fixed pattern, provides a solid sense of accom-
plishment. If the measurements are complicated, involving unusual techniques

7 The subtitle of a recent article by Krumbein (1962), "Quantification and the ad-
vent of the computer open new vistas in a science traditionally qualitative" makes
evident the overlap of our interests. Professor Krumbein's article deals explicitly with
a mechanical method of processing data; the fact that there is no mention of the use
of reasoning in testing the quality and relevance of the data to the specific issue being
investigated certainly does not mean that he thinks one whit less of the "rational
method" than I do. Similarly, I hope that nothing that I have said or failed to say is
construed as meaning that I have an aversion to mechanical methods of analyzing
data; such methods are unquestionably good if they bring out relationships not other-
wise evident, or in any other way advance the progress of the rational method of in-
vestigation. When mechanical processes replace reasoning processes, and when a num-
ber replaces understanding as the objective, danger enters.

METHODS OF INVESTIGATION IN GEOLOGY 161

and apparatus and a special jargon, they give the investigator a good feeling
of belonging to an elite group, and of pushing back the frontiers. Presentation
of the results is simplified by use of mathematical shorthand, and even though
nine out of ten interested geologists do not read that shorthand with ease, the
author can be sure that seven out of the ten will at least be impressed. It is
an advantage or disadvantage of mathematical shorthand, depending on the
point of view, that things can be said in equations, impressively, even arro-
gantly, which are so nonsensical that they would embarrass even the author
if spelled out in words.

As stated at the outset, the real issue is not a matter of classical geologic
methods versus quantification. Geology- is largely quantitative, and it is rapidly
and properly becoming more so. The real issue is the rational method versus
the empirical method of solving problems; the point that I have tried to make
is that if the objective is an understanding of the system investigated, and if
that system is complex, then the empirical method is apt to be less efficient
than the rational method. Most geologic features ledges of rock, mineral
deposits, landscapes, segments of a river channel present an almost infinite
variety of elements, each susceptible to many different sorts of measurement.
We cannot measure them all to any conventional standard of precision blind
probing will not work. Some years ago (1941) I wrote that the "eye and brain,
unlike camera lens and sensitized plate, record completely only what they
intelligently seek out." Jim Gilluly expresses the same thought more succinctly
in words to the effect that most exposures provide answers only to questions
that are put to them. It is only by thinking, as we measure, that we can avoid
listing together in a field book, and after a little while, averaging, random
dimensions of apples and oranges and apple crates and orange trees.

Briefly, then, my thesis is that the present swing to the quantitative in geology-,
which is good, does not necessarily and should not involve a swing from the
rational to the empirical method. Fm sure that geology' is a science, with
different sorts of problems and methods, but not in any sense less mature than
any other science; indeed, the day-to-day operations of the field geologist are
apt to be far more sophisticated than those of his counterpart the experi-
mentalist in physics or chemistry. And I'm sure that anyone who hires out
as a geologist, whether in practice, or in research, or in teaching, and then
operates like a physicist or a chemist, or, for that matter, like a statistician or an
engineer, is not living up to his contract.

The best and highest use of the brains of our youngsters is the working out
of cause and effect relations in geologic systems, with all the help they can get
from the other sciences and engineering, and mechanical devices of all kinds,
but with basic reliance on the complex reasoning processes described by Gilbert,
Chamberlin, and Johnson.

162 J. HOOVER MACKIX

REFERENCES CITED

ALBRTTTON, C. C., JR., 1961, Notes on the history* and philosophy of science (1) A con-
ference on the scope and philosophy of geology: J. Graduate Research Center,

Southern Methodist Univ., vol. 29, no. 3, pp. 188-192.
BUCHER, \V. H., 1933 3 The deformation of the earth's crust: Princeton, N. J., Princeton

Univ. Press, 518 pp.
, 1941, The nature of geological inquiry* and the training required for it: Am.

Inst. Mining Metall. Eng., Tech. Pub. 1377, 6 pp.
CHAMBERLIN, T. C., 1897, The method of multiple working hypotheses: J. GeoL, vol. 5,

pp. 837-848.

CONANT, J. B., 1951, Science and common sense: New Haven, Yale Univ. Press, 371 pp.
, 1952, Modern science and modern man: New York, Columbia Univ. Press,

111 pp.

GARRELS, R. M., 1951, A textbook of geology: New York, Harper, 511 pp.
GILBERT, G. K., 1886, The inculcation of the scientific method by example, with an

illustration drawn from the Quaternary geology of Utah: Am. J. ScL, vol. 31, (whole

no. 131), pp. 284-299.
, 1914, The transportation of debris by running water: U. S. Geol. Survey, Prof.

Paper, 86, 263 pp.
, 1917, Hydraulic-mining debris in the Sierra Nevada: U. S. Geol. Survey, Prof.

Paper 105, 154 pp.
HOLTON, G., 1952, Introduction to concepts and theories in physical science: Reading,

Mass., Addison-Wesley, 650 pp.
JOHNSON, DOUGLAS, 1933, Role of analysis in scientific investigation: Geol. Soc. Am., B.,

vol. 44, pp. 461-494.
KRUMBEIN, W. G., 1960, The "geological population" as a framework for analysing

numerical data in geology: Liverpool and Manchester Geol. J., vol. 2, pt. 3, pp. 341-

368.

, 1962, The computer in geology: Science, vol. 136, pp. 1087-1092.

KUENEN, P. H., 1959, Fluviatile action on sand, Part 3 of Experimental Abrasion:

Am. J. ScL, vol. 257, pp. 172-190.
LEOPOLD, L. B., 1953, Downstream changes of velocity in rivers: Am. J. ScL, vol. 251,

pp. 606-624.
, and MADDOCK, T., JR., 1953, The hydraulic geometry of stream channels and

some physiographic implications: U. S. Geol. Survey, Prof. Paper 252, 57 pp.
MACKIN, J. H., 1941, Drainage changes near Wind Gap, Pennsylvania; a study in map

interpretation: J. Geomorphology, vol. 4, pp. 24-53.

, 1948, Concept of the graded river: Geol. Soc. Am., B., vol. 59, pp. 463-512.

PLUMLEY, W. J., 1948, Black Hills terrace gravels; a study in sediment transport: J.

Geol., vol. 56, pp. 526-577.
RUBEY, W. W., 1938, The force required to move particles on a stream bed: U.S. Geol.

Survey, Prof. Paper 189-E, pp. 120-140.
SCHEIDEGGER, A. E., 1961, Theoretical geomorphology: Berlin, Springer- Verlag, 333 pp.

METHODS OF INVESTIGATION IN GEOLOGY 163

SIMPSON, G. G., ROE, ANNE, and LEWONTO, R. C., 1960, Quantitative zoology: New

York, Harcourt-Brace, 440 pp.
U. S. Geological Survey, 1960, Surface water supply of the United States, 1958; Part 6A,

Missouri River Basin above Sioux City, Iowa: U. S. Geol. Survey Water-Supply

Paper 1559, 434 pp.
WEISS, PAUL, 1962, Experience and experiment in biology: Science, vol. 136, pp. 468-

471.

MASON L. HILL

Richfield Oil Corporation

Role of Classification
in Geology

Goethe said that mineralogy* interested him only for two reasons: "I valued
it for its great practical utility, and then I thought to find a document elucidat-
ing the primary formation of the world, of which Werner's doctrine gave hopes."
But he went on to conclude that, "Since this science has been turned upside
down by the death of this excellent man, I do not proceed further in it, but
remain quiet with my own convictions." (Oxenford, 1882)

This statement, made in 1827, reflects Goethe's melancholy reaction to the
triumph of the Plutonists 5 fruitful classification of rocks over that of Werner's
less productive Neptunist school. The statement also illustrates the theme of
this essay, which is that modern geologic classifications must be revised as
knowledge increases, lest they too prove disenchanting.

Examples of attention to classification in geology are evidenced by an early
reclassification of sedimentary rocks (Grabau, 1904), by a more recent sym-
posium on the classification of sediments (Pettijohn, 1948; Shrock, 1948;
Krynine, 1948), and by a recent book on the classification of carbonate rocks
(Ham, 1962). A discussion of fault classifications is presented here to illustrate
the role of classification in geology, to emphasize the usefulness of rules of
nomenclature and classification, and more particularly to show the need for a
reclassification of faults in order to stimulate progress in structural geology.

Classification and Ordering

According to Webster, a classification is "a system of classes or groups, or a
systematic division of a series of related phenomena." To classify, one must
recognize differences and likenesses as well as gradations between things,
processes, and concepts. Classifications are particularly difficult in geology
because geologic things, processes, and concepts are many and complex; they
usually need to be fixed in space and time; and they commonly involve the

164

ROLE OF CLASSIFICATION IN GEOLOGY 165

terminology of other natural sciences. Furthermore, since the historical science
of geology is concerned with the antecedents of natural phenomena, the most
significant classifications are interpretive and are, accordingly, subject to error
both in formulation and application.

Although classification is usually conceived as a grouping on the basis of
differences and likenesses, much good classification is based on gradations of
characteristics. The arrangement of related phenomena in gradational series
has been called "ordering 39 (Hempel, 1952 and 1959), or the continuous
variable type of classification (Rodgers, 1950). Ordering is preferred whenever
it is a possible means of classification, because both large and small differences
can be expressed precisely and quantitatively. Ordering may be compared
to a "more-or-less," rather than a "yes-or-no" relationship, where one class
(or member) has characteristics which grade into another, instead of having
separate characteristics. These gradational characteristics may be expressed
either in a relative manner (before or after, larger or smaller, etc.) or quanti-
tatively, in which case mathematical analyses may be applied. Ordering is a
common means of classification in geology (e.g. isomorphous mineral series
allow quantitative chemical classification, or "late Eocene" provides compara-
tive age classification). Often artificial grouping is employed in a gradational
series to facilitate scientific work and communication (e.g., clay, silt, sand, and
gravel are useful classes of sediments within a gradational series of clast sizes).
Sometimes classes will be separated by nongradational differences in a classi-
fication, whereas members within these classes may vary gradationally and
thus be subject to ordering. Thus igneous and sedimentary rocks are generally
nongradational, whereas members of both classes usually vary gradationally.
Ordering is especially adaptable to expressing quantitative relationships in
mineralogy, petrology, paleontology, and other geologic fields, by graphs
showing two or more gradational variables.

Classifications of Faults

The history of geology records many examples of unproductive as well as
fruitful classifications, for the contrast between the Wernerian and Huttonian
classifications of rocks finds parallels in many less dramatic cases. Far too many
geologic terms are either so misleading or so commonly misused that they
actually retard scientific progress. The use of names which imply knowledge
that does not exist is particularly reprehensible. Thus genetic terms such as
"turbidites" and "slump structures" are often applied to features whose origins
are uncertain. This practice not only promulgates error but also discourages
further study of the objects under investigation. To the extent that the classi-
fications of faults now in use imply knowledge which usually does not exist,
these must be suspected of retarding progress in structural geology.

166 }^L\SOX L. HILL

The terminology of faulting presumably began more than 150 years ago in
the British coal mines with the words "at fault" used to express the abrupt
termination of coal seams. As experience was gained in locating the displaced
continuations of these seams, it was discovered that planar structures along
which movement apparently had occurred were responsible for abrupt separa-
tions of the coal beds. Such structures have continued to be known as faults,
and although defined as fractures along which relative movement has occurred,
the concept is not complete without at least some reference to the geometric
relations of one fault wall to the other. As geologists mapped other terrains,
faults of different characteristics were described. To distinguish between faults
of differing characteristics the names normal, reverse, thrust, and lateral were
gradually introduced, and used as a classification of faults. These terms are
not only common in English geologic literature, but they have equivalents in
most other languages; and the structures they denote are shown by conven-
tional symbols on geologic maps and sections. Unfortunately, these terms do
not mean the same things to all geologists, as demonstrated by variations in
definition revealed in glossaries, textbooks, periodicals, and in oral discussions.
For example, Beloussov (1962) gives a classification of faults based on relative
and absolute movements without describing criteria for the determination of
slip. Furthermore, the basic concept of fault separation as contrasted to slip
is not presented. In another section of the same textbook, a genetic classifica-
tion of faults is given without any indication as to how it is to be used. Concern
over classification is evidenced in a number of recent discussions including
those of Boyer and Muehlberger (1960), Crowell (1959), Kelley (1960), and
Kupfer (1960).

It is obvious that the sense of movement on faults is of paramount tectonic
importance. It is true, but not obvious, that geologists can rarely determine
fault slip (actual relative movement). It is also true, and again not generally
obvious, that geologists classify most faults on the basis of separations (apparent
relative movements) but erroneously use the terms normal, reverse, thrust, and
lateral in relation to slip. This confusion between the concepts of fault separa-
tion and fault slip, in the common usage of these terms, is the result of imperfect
classification.

Although Reid, et al. (1913) did much to standardize and establish modern
fault nomenclature, they .allowed the terms normal and reverse to mean ambigu-
ously either slip or separation. Gill (1941) in a committee effort which recog-
nized the difference between slip and separation, recommended the fault names
"normal," "reverse," "right," "left," and combinations thereof (e.g., "right
reverse fault") on the basis of relative slip, but without proposing complemen-
tary terms based on fault separations (the usual evidence for faults). Certainly
a logical and progressive fault classification, based on sound rules for good
nomenclature and classification, is long overdue.

ROLE OF CLASSIFICATION IN GEOLOGY

167

FIG. 1. Diagram of normal left-lateral fault. The dip separation A and the strike
separation B are used for the separation-based (apparent relative movement) classifica-
tion of this fault. The slip-based (actual relative movement) classification would be
used as follows, if slips of diagram were approximately determined: (1) reverse left-
lateral slip fault; (2) left-lateral slip fault; (3) normal left-lateral slip fault; (4) normal
slip fault; and (5) normal right-lateral slip fault.

Dual classification of faults. Most faults are recognized by discontinuities of
planar or tabular geologic elements. The accompanying diagram (Fig. 1)
shows that the displacement of a tabular geologic element with fixed strike
separation (B) and dip separation (A) can be accomplished by slips (1) through
(5) and thus demonstrates that separation (apparent relative movement) and
slip (actual relative movement) are different concepts. The fault slip (sense
and amount of relative movement) is usually indeterminable unless linear
geologic elements (e.g., line of intersection of bed and vein) are dislocated.
In such cases, and some other rare ones, the net slip is determinable. Therefore,
any generally useful classification of faults must be based on the geometry of the
separations (apparent relative movements) in relation to the fault surface;
but in the rare and tectonically important cases where slip can be determined,
a kinematic classification is more appropriate. These two natural geologic
situations call for two classifications of faults (Hill, 1959) as in Table 1.

These fault types are defined as follows:

Normal fault is an inclined surface along which there has been apparent relative
depression of the hanging wall.

Reverse fault is an inclined surface dipping 45, or more, along which there
has been apparent relative elevation of the hanging wall.

168

MASON L. HILL

TABLE 1

Separations

Faults

Slips

Faults

Dip separation

Normal ' Dip slip

Normal slip

(measured in

Reverse (measured in

Reverse slip

dip of fault)

Thrust dip of fault)

Thrust slip

Strike separation

Right-lateral Strike slip

Right-lateral slip

(measured in

Left-lateral (measured in

Left-lateral slip

strike of fault)

strike of fault)

Dip and strike

Named after prin- ; Oblique slip

Named after principal

separations

cipal separation

(measured in

slip component

(measured in

or appropriate

fault surface)

or appropriate

both dip and

combined term

combined term

strike of fault)

(e.g., normal

(e.g., reverse left-

left-lateral fault

lateral slip fault,

of diagram)

1 of diagram)

Thrust fault is a surface dipping less than 45 along which there has been
apparent relative movement of the hanging wall over the footwall. (Some work-
ers use this term either as synonymous with, subordinate to, or inclusive of
reverse fault, but there is a good case for reverse and thrust as equal and sep-
arate categories in order to distinguish between the ratios of apparent hori-
zontal shortening to apparent relative uplift.)

Right-lateral fault is a surface along which, as viewed in plan, the side opposite
the observer has had apparent relative movement to the right. (Note that the
single word "lateral" is not used in the dual classification of faults, whereas
right-lateral and left-lateral faults are as different from each other as normal
and reverse faults.

Left-lateral fault is a surface along which the side opposite the observer has
had apparent relative movement to the left.

Normal slip faulty reverse slip fault, thrust slip fault, right-lateral slip fault, and
left-lateral slip fault are defined as the separation-based terms above, with the
omission of the word "apparent," to show that information on actual relative
movement (slip) has been obtained.

These two sets of fault names, one classifying faults according to commonly
determined apparent relative movement (the strike and/or dip separations),
and the other by occasionally determined actual relative movement (the strike
and/or dip components of slip), are needed for comprehensive and practical
fault classification. The great advantage of the twofold classification is that
connotation of slip (actual relative movement) on a fault is avoided when only

ROLE OF CLASSIFICATION IN GEOLOGY 169

separation (apparent relative movement) is known. Although the separation-
based terms can usually be employed, the slip-based terms have greater tec-
tonic significance, and therefore should be used whenever possible.

Rules of Nomenclature and Dual Classification of Faults

Rules for providing good scientific classification must be formulated and
applied. First, however, rules of nomenclature (the system of names used in a
classification) must be considered. The following rules, in part modified from
Wadell (1938), are proposed, and the application of each to the dual classifi-
cation of faults is presented.

RULE 1. A name should be widely understood and used by scientists, or
it should be so appropriate that it is likely to become widely understood and
used.

Application. The separation-based fault names are widely used by geologists,
but too often in the sense of denoting slip. The terms wrench or transcurrent are
frequently used instead of lateral; but, hi this writer's opinion, wrench is objec-
tionable because of its dynamic implication; transcurrent improperly implies
transection of other structures; and both terms refer to slip without benefit of
complementary terms for use when only separations are known. A wider
understanding of the difference between separation and slip is required before
the names of faults can be commonly understood.

RULE 2. A name should be descriptive or explanatory. If the significance
(such as genesis) of the thing, action or concept is adequately known, a name
expressing such significance is desirable. However, if doubt of this signifi-
cance exists, or if the term may be used when doubt does exist, it is better
to use a purely descriptive name. A combined descriptive and explanatory
name is likely to be ambiguous and should therefore be avoided. If possible,
a descriptive name should be subject to operational definition (Bridgman,
1927), whereby others may verify by reobservation or experimentation.

Application. The fault names are descriptive rather than explanatory, al-
though normal is not rational, and thrust is dynamic rather than geometric.
Right- or left-lateral are particularly good descriptive words for apparent side-
wise movements, as right- or left-lateral slip are for relative sidewise move-
ments. These names also aptly express important geometric (or kinematic)
differences. Kelley (1960) and others have objected to naming faults on the
basis of separations, because these can vary along the fault (in plan or section).
Obviously in some such cases, due to complicated structural or stratigraphic
situations, the separation-based terms characterize only local segments of the
fault. But in many of these cases such variations of separation provide evidence

170 MASON L. HILL

for fault slip. For example, separations resulting from the displacement of an
anticline may clearly indicate either dip slip, strike slip, or some combination
thereof. In these cases the fault can be named according to the slip, which is
descriptive of the kinematics and thus is technically more significant than a
name based upon the separation alone.

RULE 3. If possible, a name should be a common word, with modifying
words as required; it should approximate a definition of a class, with modifiers
to distinguish between members of the class. Proper names and terms com-
bining roots of different linguistic origins are permissible only if they are
more practical. When a common or vernacular name is used in a scientific
sense, its meaning must be defined. It is better to adopt a new name that is
precisely defined than to use a common name whose meaning is inexact.

Application. The fault names are precisely defined common nouns with ap-
propriate adjectives. It has been suggested by Gill (1935) and others, that the
qualifying word "separation" be added to the fault term when slip has not
been determined. However, experience indicates that this limiting word will
seldom be appended, whereas if slip had been determined the author would be
eager to add the word "slip" to the fault name as evidence of additional in-
formation. It has been suggested by Kelley (1960) that right and left slip be
substituted for right- and left-lateral slip, respectively. If these terms are
accepted, it would be logical to change the separation-based terms to right
fault and left fault, or right-separation and left-separation faults. Furthermore,
if the term separation were used in the dual classification, it would also be logical
to refer to normal separation, reverse separation, and thrust separation faults.
These and other reasonable suggestions for fault names, including wrench and
transcurrent, need to be accepted or rejected so that one set of names can be
employed. Recommendations from a duly appointed international committee
may be required to establish standards.

RULE 4. A name should be rational and appropriate to the science
involved, although a new name should not be used if an appropriate old one
exists within the science or if an acceptable one from another science is
available. However, new and precisely defined terms are preferable to old
terms that are imprecise or commonly misused.

Application. The fault names are, in context, rational and appropriate to the
science of geology. New names are not used in the dual classification of faults,
although more precise and restricted definitions are employed. If the names
were not in common usage, new names for both separation-based and slip-
based classifications might be preferred.

ROLE OF CLASSIFICATION IN GEOLOGY 171

RULE 5. A foreign term should be used when appropriate, but unless it
is a common word it should be used in its original rather than in its translated
form.

Application. Foreign names are not used in the classification, although some
are properly used in fault terminology.

RULE 6. Symbolic and/or mnemonic terms may be used if properly
defined, and if more practical than descriptive terms.

Application. Since the terms are few and short, symbolic or mnemonic terms
are not justified.

RULE 7. The same thing should not be given two different names, nor
should two different things be given the same name. It is, however, less
objectionable for a thing to have two names than for two different things to
have the same name.

Application. The fault names of the dual classification are not ambiguous.
If slip is determined, only the slip-based nomenclature is used, even though
the fault may have previously been classified according to separation. The
common usage of a single set of fault terms in a slip sense when only separation
is known is ambiguous, and has conspired to prevent a search for criteria
determining fault slip. For example, hi common usage "reverse foul? 9 can
mean either that reverse slip or that reverse separation is known. This usage
has also encouraged ill-founded tectonic interpretations, and has generally
retarded progress in structural geology. Use of the dual nomenclature can
preclude ambiguity and encourage determination of fault slip.

RULE 8. A name should represent a group of things, processes, or concepts
and, if possible, should also be a part of a greater group (e.g. granite repre-
sents a mineralogical and textural group of rocks and it is also a class of
igneous rocks).

Application. The names of the dual classification represent groups of faults,
and faults are only one group of earth structures.

Rules of Classification and Dual Classification of Faults

RULE A. A classification (including its nomenclature) should be appro-
priate to the knowledge of the subject, and to the purposes for which it is
devised.

Application. The dual classification of faults is appropriate for current knowl-
edge of faulting and for the purposes for which it is devised. There are other

172 MASOX L. HILL

ways of classifying faults, and for special purposes some of these may be better
than the dual classification. For example, bedding-plane, dip, strike, and
oblique faults, or longitudinal and transverse faults, or epi-anticlinal faults
provide grouping according to associated structures. But since such classi-
fications are not based on the characteristics of the faults themselves, they
cannot be considered as significant classes of faults. Although fault classifica-
tions based on presumed origins (Clark, 1943) or on presumed stress fields
(Anderson, 1951) have had considerable acceptance, it appears that genetic
or dynamic classifications are either too theoretical or too subjective for present
knowledge of structural geology- (Hill, 1947). The dual classification allows
appropriate classification of most faults on the basis of apparent relative move-
ment (separation), and in addition provides a classification for faults when
actual relative movement (slip) is determined.

RULE B. A classification should be comprehensive and exclusive. Thus
the categories of the classification should be collectively complete and indi-
vidually separate.

Application. The classification is comprehensive and exclusive. All recognized
cases of fault separation and slip are included, and the five rational categories
in each (normal, reverse, thrust, and right- and left-lateral) are mutually
exclusive.

RULE C. A classification should be flexible. Provision should be made
for intercalation of new terms and for expansion of existing terms. Classifi-
cation based on ordering of gradational characteristics should be used where
possible, because greater precision in differentiation is allowed without
preventing broad grouping.

Application. The classification is such that all the principal fault types are
named, all combination types can be expressed (e.g. normal left-lateral fault),
and all possible components of separation or slip are included. No expanded
connotation of the terms can be conceived until dynamic and/or genetic fault
characteristics are commonly recognized. Ordering of all gradations of strike
and/or dip of the fault, all variations of separation, and all variations in sense
of slip may be expressed within the categories of the dual classification. Al-
though, for example, normal and reverse (or normal-slip and reverse-slip)
faults are gradational through the vertical, ordering across a vertical surface
is not good classification because this dip divides apparent (or actual, if slip is
known) horizontal extension from shortening across the fault. On the other
hand, ordering of fault dip within one class is significant in order to obtain the
proportion of relative uplift to either extension or shortening.

ROLE OF CLASSIFICATION IN GEOLOGY 173

RULE D. A classification should be either descriptive or theoretical.
Combinations of descriptive and theoretical elements lead to confusion. As
the science progresses theoretical classifications become more useful.

Application. The dual classification of faults is based entirely on the descrip-
tive elements of separation (geometric) or slip (kinematic), although it is pos-
sible that future advancements in the knowledge of faulting may allow dynamic
or even genetic classifications.

Conclusions

In geology, as in all other sciences, progress is facilitated by precise nomen-
clature and logical classification. All geologic classifications should be scruti-
nized in the light of current knowledge. To stimulate advancements in their
science, geologists must be as willing to revise their classifications as they are to
make new observations and new interpretations. The rules proposed here for
naming and classifying geologic phenomena are thought to be useful, although
they may not be complete or mutually exclusive, are not of equal importance
or scope, and could be expressed in different terms.

Present classifications of faults have retarded progress in structural geology
because the names commonly given to faults so often falsely imply knowledge
of fault slip. The dual classification of faults recommended here is analogous
to the dual classification of strata into rock-stratigraphic and time-stratigraphic
categories. In stratigraphy progress was stimulated by a classification which
distinguished between lithologic and chronologic units. A classification which
distinguishes between the concepts of fault separation and fault slip might
likewise stimulate progress in structural geology.

REFERENCES CITED

ANDERSON, E. M., 1951, The dynamics of faulting and dyke formation with applications

to Britain, 2nd ed.: London, Oliver and Boyd, 206 pp.
BELOUSSOV, V. V., 1962, Basic problems in geotectonics (Am. Geol. Inst. translation):

New York, McGraw-Hill, 816 pp.
BOYER, R. E., and MUEHLBERGER, W. R., 1960, Separation versus slip: Am. Assoc.

Petroleum Geol., B., vol. 44, pp. 1938-1939.

BRIDGMAN, P. W., 1927, The logic of modern physics: New York, Macmillan, 228 pp.
CLARK, S. K., 1943, Classification of faults: Am. Assoc. Petroleum Geol., B., voL 27,

pp. 1245-1265.
CROWELL, J. C., 1959, Problems of fault nomenclature: Am. Assoc. Petroleum Geol.,

B., vol. 43, pp. 2653-2674.

174 MASOX L. HILL

GILL, J. E., 1935, Normal and reverse faults: J. GeoL, vol. 43, pp. 1071-1079.
, 1941, Fault nomenclature: Roy. Soc. Canada, Tr., 3rd ser., vol. 35, sec. 4,

pp. 71-85.
GRABAU, A. W., 1904, On the classification of sedimentary rocks: Am. Geologist, vol. 33,

pp. 228-247.
HAM, W. E., ed., 1962, Classification of carbonate rocks, a symposium: Am. Assoc.

Petroleum GeoL, Mem. 1.
HEMPEL, C. G., 1952, Fundamentals of concept formation in empirical science: Internat.

Encyclopedia of Unified Sciences, vol. 2, no. 7, Chicago, Univ. Chicago Press, 93 pp.

, 1959, Some problems of taxonomy: unpublished manuscript.

HILL, M. L., 1947, Classification of faults: Am. Assoc. Petroleum GeoL, B., vol. 31,

pp. 1669-1673.
, 1959, Dual classification of faults: Am. Assoc. Petroleum GeoL, B., vol. 43,

pp. 217-221.

, 1960, Fault symbols: GeoTImes, vol. 5, no. 3, pp. 23-24.

KELLEY, V. C., 1960, Slips and separations: GeoL Soc. Am., B., vol. 71, pp. 1545-1546.
KRYNINE, P. D., 1948, The megascopic study and field classification of sedimentary

rocks: J. GeoL, vol. 56, pp. 130-165.
KUPFER, D. H., 1960, Problems of fault nomenclature: Am. Assoc. Petroleum GeoL, B.,

vol. 44, pp. 501-505.

OXENFORD, JOHN, 1882, Conversations of Goethe: London, George Bell and Sons.
PETTIJOHN, F. J., 1948, A preface to the classification of sedimentary rocks: J. GeoL,

vol. 56, pp. 112-117.
REID, H. F., DAVIS, W. M., LAWSON, A. C., and RANSOME, F. L., 1913, Report of the

Committee on the Nomenclature of Faults: GeoL Soc. Am., B., vol. 24, pp. 163-183.
RODGERS, JOHN, 1950, The nomenclature and classification of sedimentary rocks: Am*

J. Sci., vol. 248, pp. 297-311.

SHROCK,R.R., 1948, A classification of sedimentary rocks: J. GeoL, vol. 56, pp. 118-129.
WADELL, HAKON, 1938, Proper names, nomenclature and classification: J. GeoL,

vol. 46, pp. 546-568.

CHARLES A. ANDERSON

U. S. Geological Survey

Simplicity in

Structural Geology 1

In 1951, I made the statement that "Until more precise correlations of the
older Precambrian rocks can be made, based on radioactivity or other methods,
the simplest explanation is that only one period of orogeny, corresponding to
Wilson's Mazatzal Revolution, has occurred in Arizona during early Pre-
cambrian time." (Anderson, 1951, p. 1346; italics added) This explicit refer-
ence to the use of simplicity in correlating structural events and reconstructing
geologic history led the Chairman of the Anniversary Committee to ask for an
essay on simplicity in structural geology.

Principle of Simplicity

The principle of simplicity has been called Occam's (or Ockham's) razor,
the principle of parsimony, or the principle of economy. Allusions to simplicity
in the literature are innumerable and varied in intent and nuance. A revival
of interest in this principle among philosophers of science has been partly
inspired by the work of Nelson Goodman.

All scientific activity amounts to the invention of and the choice among
systems of hypotheses. One of the primary considerations guiding this
process is that of simplicity. Nothing could be much more mistaken than
the traditional idea that we first seek a true system and then, for the sake of
elegance alone, seek a simple one. We are inevitably concerned with sim-
plicity as soon as we are concerned with system at all; for system is achieved
just to the extent that the basic vocabulary and set of first principles used in
dealing with the given subject matter are simplified. When simplicity of

1 1 wish to thank my two colleagues, James Gilluly and Walter S. White, for helpful
and critical comments in their review of this essay.

175

1 7 6 CHARLES A. ANDERSON

basis vanishes to zero that is, when no term or principle is derived from
any of the others system also vanishes to zero. Systematization is the same
thing as simplification of basis. (Goodman, 1958, p. 1064}

William of Occam, known as Doctor Invincibilis and Venerabihs Inceptor, was
born around 1300 and became a member of the Franciscan order while still a
youth. He was an intellectual leader in the period that saw the disintegration
of old scholastic realism and the rise of theological skepticism. The famous
dictum attributed to him, "Entia non sunt multiplicanda praeter necessitate" has
appeared in nearly every book on logic from the middle of the nineteenth
century*. It is doubtful that the "Invincible Doctor" used these words (Thor-
burn, 1918), but he did use similar words such as Frustrafit per plura quod potest
fieri per pauciora (Laird, 1919, p. 321). The "razor" is commonly used now
without special reference to the scholastic theory- of entities, and Laird believes
that the precise form in which Occam stated it is irrelevant.

Russell (1929, p. 113) states that Occam's razor, "entities are not to be multi-
plied without necessity," is a maxim that inspires all scientific philosophizing,
and that in dealing with any subject matter, one should find out what entities
are undeniably involved, and state everything in terms of these.

The concept of simplicity is a controversial topic in the philosophy of science
according to Kemeny (1953, p. 391 ). One school believes it involves an assump-
tion about the simplicity of nature, whereas others justify it as a matter of con-
venience, "a labor-saving device." Jevons (1883, p. 625) objected to the gen-
eralization that the laws of nature possess the perfection which we attribute to
simple forms and relations, and suggested that "Simplicity is naturally agree-
able to a mind of limited powers, but to an infinite mind all things are simple."

Mill (1865, p. 461) questioned Hamilton's belief that "Nature never works
by more complex instruments than are necessary*," stating that "we know
well that Nature, in many of its operations, works by means which are of
a complexity so extreme, as to be an almost insuperable obstacle to our in-
vestigations." Mill (1865, p. 467) believed that we are not justified in reject-
ing an hypothesis for being too complicated, but "The 'Law of Parcimony 5
needs no such support; it rests on no assumptions respecting the ways or pro-
ceedings of Nature. It is a purely logical precept; a case of the broad practical
principle, not to believe anything for which there is no evidence . . . The
assumption of a superfluous cause, is a belief without evidence." Mill (ibid.)
emphasizes that the principle which forbids the assumption of a superfluous
fact, forbids a superfluous law and "The rule of Parcimony, therefore, whether
applied to facts or to theories, implies no theory concerning the propensities or
proceedings of Nature."

Feuer (1957, p. 121), like Mill, emphasized that the scientific principle of
simplicity does not rest on the assumption that the laws of nature are simple,

SIMPLICITY IN STRUCTURAL GEOLOGY 177

and he pointed out that the simplicity of nature has had a long philosophical
history which he would call the metascientific principle of simplicity, to distinguish
it from the scientific methodological principle of Occam's razor. Verifiability
is the important element in Occam's razor; the principle of simplicity is thus a
straightforward basis for rejecting theories if they are unverifiable (Feuer,
1957, p. 115). The verified theory is the simplest because every unnecessary'
component is an unverified item.

Demos (1947) expresses some skepticism about the use of simplicity and
suggests that the scientific philosopher tries to evade the charge of fallacious
reasoning by introducing the principle of simplicity, thereby enabling him to
choose among the several theories consistent with the observed facts. The
scientist then selects the theory that explains the greatest number of phenomena
with the fewest assumptions. Bridgman (1961, p. 10) regarded Occam's razor:

... as a cardinal intellectual principle, ... I will try to follow it to the ut-
most. It is almost frightening to observe how blatantly it is disregarded in
most thinking ... To me it seems to satisfy a deep-seated instinct for intel-
lectual good workmanship. Perhaps one of the most compelling reasons for
adopting it is that thereby one has given as few hostages to the future as
possible and retained the maximum flexibility for dealing with unanticipated
facts or ideas.

Simplicity and Geology

A recent issue of "Philosophy of Science" contains a symposium on simplicity,
but only one paper mentions its application to geology. Barker (1961, p. 164)
revives the old problem of the meaning of fossils: are they remains of organisms
that actually existed on earth millions of years ago, or were they placed there
by the Creator to test our faith? Barker concluded that unless there is inde-
pendent evidence in favor of a Creator, the simple theory is that plants and
animals existed in the past in circumstances similar to those in which we find
them today. This leads to uniformitarianism, a topic discussed elsewhere in
this volume.

An excellent geologic example of simplicity is given by Woodford (1960) in
discussing the magnitude of strike slip on the San Andreas fault. In 1906, the
right-lateral slip along this fault was 22 feet in central California, and in south-
ern California offset streams indicate right-lateral movement of thousands of
feet during Quaternary time. But for pre-Quaternary movements, it is necessary
to distinguish between separation and slip. A structurally complex succession
of granodiorite, Paleocene, and Miocene rocks north of the San Gabriel
Mountains is offset in a way that seems to require 30 miles of right-lateral slip
since the middle Tertiary. Displacements may have been even greater (a range
from 160 to 300 miles has been suggested), but Woodford prefers a working

178 CHARLES A. ANDERSON

hypothesis that includes some dip slip and so he limits strike slip on the San
Andreas fault to 30 miles, right lateral. "The tentative choice of short slips, if
these will do the business, is an example of the use of the principle Disjunctiones
minimae> disjunction's optimae. This rule may be considered a quantitatively
parsimonious relative of Ockham's law: Entia non sunt multiplic anda praeter neces-
sitatem." (Woodford, 1960, p. 415)

The preparation of a geologic map is the essential first step in structural
geology, and one of the first steps is the building up of the stratigraphic se-
quence. The law of superposition is vital to the success of this study. Even
where we have "layer-cake" stratigraphy, Albritton (1961, pp. 190-191) has
pointed out that it is not clear how the principle of simplicity operates in
geology:

. . . [given] two nearby mesas of three formations conformably arranged in
similar sequence from bottom to top. Without evidence to the contrary,
most stratigraphers would recognize only three formations in all, perhaps on
the ground that it is in vain to do with more what can be done with fewer.
But if a three-formation column is simpler than a [six-formation] column,
would it not be simpler still to lump the three formations into one group,
and then have a single entity?

Of course the answer is that to do so is to lose information. The purpose of
the geologic study has an important bearing on the choice of stratigraphic units.
Fundamentally, selection is made to focus attention on the environment of
deposition of the sediments, to indicate the various stages in the geologic history.
For structural interpretations, delineation of thin units may help the geologic
map to elucidate the structure. The objective is to use the map as a means of
showing as many as possible of the elements that bear on the geologic history
and structure, and something about the basis on which these elements have
been verified.

In regions of complex structure, particularly if the rocks are nonfossiliferous
and folded isodinally, stratigraphy and structure are determined concurrently
by mapping distinctive lithologic units whether they are beds, zones, or forma-
tions. Structural elements and data on the direction that the tops of beds are
now facing are diligently searched for in all exposures. In this manner, the
stratigraphy and structure unfold together, and the final interpretation results
from the integration of both. The interpretation may be complex, but the
principle of simplicity will be followed if no unverifiable facts are essential to
the interpretation. In actual practice there are few situations in which a geo-
logic interpretation does not require some unverified assumptions, and in
general the use of the principle of simplicity involves the acceptance of the
interpretation that has the maximum of verifiable facts and minimum of
assumptions.

SIMPLICITY IN STRUCTURAL GEOLOGY 179

Prediction is an important facet of structural geology; in mapping, a field
geologist commonly predicts what will be found on the next ridge, valley, or
mountain range. In a sense, this is a field test of the interpretations developing
in the mind of the geologist. No doubt it is a frequent experience with a
geologist, mapping in a region of complex geology, to find his predictions
erroneous; the geology may be more complex than the interpretation of the
moment. But this is a part of the accumulation of field data and in no way
conflicts with the use of simplicity in the final interpretation. Prediction is the
end product of many studies in structural geology in proposing exploration
programs, and in the search for new mineral deposits and petroleum accumu-
lations.

Many examples could be cited where early expositions of the geologic history
of a particular region are less complex than later explanations based on addi-
tional field studies. Probably most geologists would accept as axiomatic that
new information leads to a more complex story. In a sense, it is a mark of
progress as we build upon the experience of those who worked before us on
similar problems. This is to be expected, particularly in regions where the
rocks have been acutely deformed by past tectonic activity. The geologic
history becomes more complex as we build up a storehouse of "verifiable
elements," even though each succeeding historical account does not introduce
entities beyond necessity.

In regions where heavy vegetation and thick soil cover the rocks, it is a
time-consuming process to assemble the facts that are needed to reconstruct
the story of the stratigraphy and structure. The early interpretations are likely
to be simple because of meager data. Trenching, drill holes, and painstaking
studies of the saprolites may in time add to the verifiable elements to give a
more complete and more complex history.

Older Precambrian in Arizona

I would like to discuss in more detail the older Precambrian geology in
Arizona as an example of the workings of the principle of simplicity in struc-
tural geology. By 1951, sufficient geologic mapping had been done in the
older Precambrian rocks of the Grand Canyon, Globe-Miami, Mazatzal
Mountains, Little Dragoon Mountains, Bagdad, and Prescott-Jerome areas
to indicate that only one period of orogeny, followed by the intrusion of granitic
rocks, could be recognized in each of these areas.

The natural temptation is to assume that the orogenies in these five
separate areas occurred at the same time, particularly because of the general
parallelism of the folds where the trends were determined, and Wilson
(1939) has termed this probable widespread erogenic disturbance, the
Mazatzal Revolution. From a purely academic view, one might question

180 CHARLES A. ANDERSON

this conclusion, for it is well known that the Precambrian covers an immense
period of time, and it would be surprising if only one period of orogeny
occurred in Arizona during the early Precambrian time. Hinds (1936) has
suggested that two periods of orogeny and two periods of granitic invasion
occurred in Arizona prior to the deposition of the Younger Precambrian
Grand Canyon series and Apache group, the Mazatzal quartzite marking
the period of sedimentation between these orogenies. Because no positive
angular unconformities have been found between the Mazatzal quartzite
and Yavapai schist, some doubt exists regarding the validity of this older
period of orogeny and granitic invasion. (Anderson, 1951, p. 1346; there-
upon followed the sentence quoted in the introductory* paragraph.)

Philip M. Blacet of the U.S. Geological Survey has recently mapped, south
of Prescott, a basement of granodiorite gneiss older than the Yavapai Series.
This gneiss underlies a basal conglomerate containing angular blocks of the
granodiorite and abundant well-rounded boulders of aplite. The basal con-
glomerate grades upward into feldspathic sandstone, gray slate, pebble con-
glomerate, and beds of rhyolitic tufiaceous sandstone (now recrystallized to
quartz-sericite schist; of the Texas Gulch Formation, described by Anderson
and Creasey (1958, p. 28) as possibly the oldest formation in the Alder Group
of the Yavapai Series. Similar slate, pebble conglomerate, and tuffaceous
sandstone occur in the type section of the Alder Group hi the Mazatzal Moun-
tains (Wilson, 1939, p. 1122) and were deformed during Wilson's Mazatzal
Revolution. The unconformable relation between the Texas Gulch Formation
and the granodiorite gneiss south of Prescott is important as proving the
existence of a granitic rock older than the Yavapai Series, and therefore older
than the granitic rocks intruded during the Mazatzal Revolution. Therefore
two periods of granitic intrusion in the older Precambrian of Arizona are
proved by normal stratigraphic relations, superposition, and transgressive
intrusive contacts. Hinds was correct in suggesting the two periods, but he
placed his second period after the Mazatzal Revolution rather than before.
He did not have verifiable data to support his conclusion, and following the
principle of simplicity, his contribution had to be ignored.

Progress is being made in the use of radiometric measurements to correlate
rocks and structural events in the older Precambrian of Arizona. Some of these
data are shown in Table 1 ; the ages indicated for the gneisses, granites, and
pegmatites from which mica samples were collected range from around 1200
to around 1500 million years (m. y.).

Mica from the pre- Yavapai granodiorite gneiss south of Prescott gave K-Ar
and Rb-Sr measurements indicating an age of about 1250 m. y. (Carl Hedge,
written communication). Measurements of the isotopes of lead in the zircon
from the granodiorite gneiss indicate a minimum age of 1700 m. y. (E. J. Catan-
zaro, oral communication). These data indicate that the mica in the granodi-

SIMPLICITY IN STRUCTURAL GEOLOGY 181

TABLE 1

RADIOMETRIG AGES OF ROCKS
FROM THE OLDER PRECAMBRIAN OF ARIZONA

. Ages in million years

Sample locations*

F K-Ar Rb-Sr

1.

Gneiss, Grand Canyon

1390

1370

2a.

Lawler Peak granite, Bagdad

1410

1390

2b.

Pegmatite in Lawler Peak granite

1410

1500

3.

Pegmatite, Wickenburg

1160

1300

4.

Pegmatite in Vishnu schist,

1550

Grand Canyon

1530

5.

Migmatite zone in Vishnu schist,

Grand Canyon

1390

6.

Granite near Valentine

1300

7.

Diana granite, Chloride

1350

8.

Chloride granite, Chloride

1210

9.

Oracle granite, Oracle

1450

* Samples 1 through 3 are from Aldrich, Wetherill, Davis, 1957, p. 656,
and samples 4 through 9 are from Giletti and Damon, 1961, p. 640.

orite gneiss recrystallized during the deformation of the Yavapai Series, cor-
responding in a general way to the time Aldrich, Wetherill, and Davis (1957)
have called the 1350-m.y. period of granitic rocks. The zircon gives an older
age, more in keeping with the stratigraphic relations. It should be noted that
Silver and Deutsch (1961) have reported an age of 1650 m. y. for zircon from
a granodiorite in southeastern Arizona (Cochise County).

It is tempting to assume that the 1350-m.y. period corresponds to the
Mazatzal Revolution; unfortunately no reliable radiometric dates have been
obtained from granitic rocks clearly related to the Mazatzal orogeny, that is,
from the Mazatzal Mountains or adjacent areas. The available age data
clearly demonstrate the need for systematic work, for there is much to be
learned from radiometric measurements of the Precambrian rocks in Arizona.
I predict that such studies will ultimately show that the structural history is
more complex than can be documented from present data.

Most structural geologists would infer that there are, in Arizona, meta-
morphic rocks older than the granodiorite gneiss south of Prescott and that such
rocks were deformed prior to or during the intrusion of the granodiorite made
gneissic during the deformation of the Yavapai Series. Using the principle of
simplicity, we can say with assurance that the simplest explanation in 1962 is

182 CHARLES A. ANDERSON

that there were at least two periods of orogeny and granitic intrusion in the
older Precambrian history of Arizona. As more verifiable elements are dis-
covered, the story may well become even more complex.

Concluding Statement

Much of my discussion has been limited to the use of the principle of sim-
plicity in explanatory' or interpretive aspects of structural geology rather than
in developing theories or laws. It is appropriate to refer to Mario Bunge, who
raises doubts about simplicity in the construction and testing of scientific
theories. A theory must at least be consistent with the known facts and should
predict new and unsuspected facts (Bunge, 1961, p. 133). Bunge (1961, p. 148)
believes that simplicities are undesirable in the stage of problem finding, but
desirable in the formulation of problems, and much less so in the solution of
problems. His advice is that "Ockham's razor like all razors must be
handled with care to prevent beheading science in the attempt to shave off
some of its pilosities. In science, as in the barber shop, better alive and bearded
than dead and cleanly shaven."

It seems to me that when a structural geologist is formulating explanatory
hypotheses, the principle of simplicity should not restrict his imagination;
complex hypotheses may stimulate and guide the work toward new and differ-
ent data. For this phase of a study, Bunge has made an excellent point; it is
only in the final selection of the hypotheses that the assortment should be pared
by Occam's razor.

REFERENCES CITED

ALBRTTTON, C. C., JR., 1961, Notes on the history and philosophy of science. (1) A con-
ference on the scope and philosophy of geology: J. Graduate Research Center,
Southern Methodist Univ., vol. 29, no. 3, pp. 188-192.

ALDRICH, L. T., WETHERELL, G. W., and DAVIS, G. L., 1957, Occurrence of 1350
million-year-old granitic rocks in western United States: Geol. Soc. Am., B., vol. 68,
pp. 655-656.

ANDERSON, C. A., 1951, Older Precambrian structure in Arizona: Geol. Soc. Am., B.,
vol. 62, pp. 1331-^6.

and CREASEY, S. C., 1958, Geology and ore deposits of the Jerome area, Yavapai

County, Arizona: U. S. Geol. Survey, Prof. Paper 308, 185 pp.

BARKER, S. F., 1961, On simplicity in empirical hypotheses: Phil. ScL, vol. 28, pp. 162-
171.

BRIDGMAN, P. W., 1961, The way things arc: New York, Viking (Compass Books Edi-
tion) 333 pp.

BUNGE, MARIO, 1961, The weight of simplicity in the construction and assaying of
scientific theories: Phil. Sci., vol. 28, pp. 120-149.

SIMPLICITY IN STRUCTURAL GEOLOGY 183

DEMOS, RAPHAEL, 1947, Doubts about empiricism: Phil. Sci., vol. 14, pp. 203-218.
FEUER, L. S., 1957, The principle of simplicity: Phil. Sci., vol. 24, pp. 109-122.
GILETTI, B. J. and DAMON, P. E., 1961, Rubidium-strontium ages of some basement

rocks from Arizona and northwestern Mexico: Geol. Soc. Am., B., vol. 72, pp. 639-

644.

GOODMAN, NELSON, 1958, The test of simplicity : Science, vol. 128, pp. 1064-1069.
HINDS, N. E. A., 1936, Uncompahgran and Beltian deposits in western North America:

Carnegie Inst. Washington, Pub. 463, pp. 53-136.

JEVONS, W. S., 1883, The principles of science: London, MacMillan, 786 pp.
KEMENY, J. G., 1953, The use of simplicity in induction: Phil. Rev., vol. 62, pp. 391-408.
LAIRD, JOHN, 1919, The law of parsimony: The Monist, vol. 29, p. 321-344.
MILL, J. S., 1865, An examination of Sir William Hamilton's philosophy: London,

Longmans, Green, 561 pp.
RUSSELL, BERTRAND, 1929, Our knowledge of the external world: New York, W. W,

Norton, 268 pp.
SILVER, L. T. and DEUTSCH, SARAH, 1961, Uranium-lead method on zircons: New

York Acad. Sci., Ann., vol. 91, pp. 279-283.

THORBURN, C. C., JR., 1918, The myth of Occam's Razor: Mind, vol. 27, pp. 345-353.
WILSON, E. D., 1939, Pre-Cambrian Mazatzal Revolution in central Arizona: Geol. Soc.

Am., B., vol. 50, pp. 1113-1164.
WOODFORD, A. O., 1960, Bedrock patterns and strike-slip faulting in southwestern

California: Am. J. Sci., vol. 258A, pp. 400-417.

LUNA B. LEOPOLD
WALTER B. LANGBEIN

U.S. Geological Survey

Association and
Indeterminacy
in Geomorphology

You find a rock. It looks like an ordinary piece of flint, broken and rough.
On a part of it is a patina whose soft grey color contrasts with the shiny brownish
surfaces of conchoidal fracture. You could have found this rock in nearly any
kind of an environment almost anyplace in the world. There is nothing dis-
tinctive about it.

You hand this same piece of rock to a colleague and ask what he can make
of it. He considers it soberly before he says, "You know, that could be an
artifact." There springs to mind then a picture of a primitive man, squatting
barefoot before a fire warming his hands. The firelight casts his shadow
against the cliff below which he crouches.

The difference between the reaction before and after the passing thought
that this might indeed be the tool of ancient man is the difference between
mild disinterest and a kaleidoscope of mental pictures. This difference reflects
differences in the associations of thoughts.

The present essay is concerned with how associations are used in geologic
reasoning, and then with certain philosophic considerations which seem to be
influencing the methodology and direction of geomorphology.

When you picked up the piece of flint the associations which flashed through
your mind were specific to the limits of your knowledge regarding the object
itself. This stone was unusual only in that it appeared to have been worked by
human hands. The mental pictures which were projected by the thought
process stemmed only from an intellectual interest. The specimen itself is
valueless. If, however, the rock had been a sample of ore, the chain of thought
might have led to interest of quite a different kind. Our everyday experience

184

ASSOCIATION AND INDETERMINACY IN GEOMORPHOLOGY 185

in geology emphasizes that the purposes of this branch of natural science are
twofold intellectual and utilitarian being constituted of the two principal
elements which generally tend to stimulate the mind of man.

In geomorphology, as in other branches of science, mental pictures depicting
associations in the natural world have an intrinsic value which stems from the
wonderment that a knowledge of nature seems to produce nearly uniformly
among thoughtful human beings. But associations in the natural world are
not only objects of interest in themselves; they are also tools of the art.

The association of different observations is a form of logic. What is here
called "association" might be viewed by some merely as another word for
reasoning. But this type of reasoning which is used in geology is so extensively
elaborated that it bears but little resemblance to mathematical logic, even if
the logician may be able to discern in geologic reasoning the same precepts
and, indubitably, the same methods which constitute the bases for any kind of
logical reasoning. If the reader, then, wishes to equate the word "association"
as used here with logic or with reasoning, we pose no objection, but it is the
basis for this reasoning that is here being examined.

The simplest and most fundamental type of association deals with the process
acting. When one observes in an outcrop a uniformly bedded sandstone he
associates this with his general knowledge of the way in which sand may be
deposited. A sand deposit usually implies that there was a source of quartz
materials, a process by which these materials were reduced to relatively uni-
form size and sorted, and a physical situation leading to progressive accumula-
tion of the materials in a depositional environment. The outcrop is inter-
preted, then, by means of a general knowledge of the processes of weathering
and subsequent transportation by water or wind.

In contrast to the observations of materials in a vertical section, another line
of associations relates a feature of the landscape to particular processes. The
occurrence of an alluvial fan at the mouth of a canyon is interpretable in terms
of the form and location of the materials, in this case both indicating that the
sediment making up the fan had its source in the canyon and that it was trans-
ported from there to its present position, presumably by water or by gravita-
tional flow lubricated by water.

Implicit in the utilization of associations is the principle of uniformitarian-
ism: geologic processes presently observed are presumed to be the same as
those operating throughout geologic time. The association of a cropping of
uniformly bedded sandstone with presently observed conditions under which
sand may be so deposited stems from the assumption that processes presently
observed are the same ones that operated in the distant past.

The concept of association goes far beyond a principle even so general as
that of uniformity. That principle in itself does not necessarily suggest sequen-
tial operations, nor does it treat of the relationship between individual observa-

186 LEOPOLD AND LAXGBEIX

tions and the generality to which these observations may be applicable. For
example, in the sample case cited, let it be supposed that the sandstone is
transected by a dike of igneous material. The knowledge of process leads to
the conclusion that the sand must have been deposited at a time previous to
that in which the igneous material was intruded.

The idea of uniformitarianism does not in itself deal with time relationships.
The geologist studies the bones of a dinosaur. In the same formation where
the bones were discovered the footprints of a beast are found preserved as casts.
The bones of the feet can be compared with the footprint and, let us say, the
print seems to have been made by the animal whose bones are now fossilized.
There is nothing at the present time quite like this creature, and it is by the
use of association rather than by reasoning stemming from uniformitarianism
that the bodily form of the dinosaur can be shown to be compatible with the
casts of his footprints. It would seem, then, that the use of associations provides
an indispensable extension to uniformitarianism in geologic reasoning.

Whole fields of geology, particularly paleontology, are based more on asso-
ciation than on any principle which relates presently observable processes to
those which occurred in previous epochs. Interpretations must be made of
phenomena unlike any known to occur under present conditions. This implies
that the concept of association is of no lower an order of generality than is the
principle of uniformitarianism.

To summarize, then, geologic reasoning is based on a logic called here the
use of associations. Associations are useful in four different ways. First, par-
ticular associations may indicate the sequence of events in time. Second, an
association found locally may indicate a general relation having limits far
beyond the immediate locality or scope of the observation. Third, a particular
association may be indicative of the processes acting. Fourth, synthesis of a
variety of observations is, in effect, a broadening of the scope of associations
considered in a given context.

Generalization may be thought of as a synthesis of individual bits of knowl-
edge into a broader framework, but synthesis is merely the broadening of a
context of association. The number of associations which are involved in a
particular thought process is possibly one measure of the degree to which syn-
thesis is achieved. Thus the use of association, much in the manner indicated
by the simple examples cited above, constitutes the methodology of synthesis,
or integration. From this point of view the utilization of the concept of associa-
tion represents one of the fundamental bases of geology.

In the inductive method, the purpose of describing a phenomenon may be,
for example, to eliminate extraneous details to see what, on the average, is
the pattern represented by the data. The generalized description may be either
quantitative or qualitative. The question of what data should be included

ASSOCIATION AND INDETERMINACY IX GEOMORPHOLOGY 187

would be determined primarily by the question asked rather than by an
a priori determination of whether the data apply to the generalized description
required. As many cases as possible would be studied to see what patterns are
displayed among the examples. Whether quantitative or qualitative, the search
for patterns in information is essentially inductive.

The difference between inductive and deductive approaches does not lie in
the presence or absence of a working hypothesis or multiple hypotheses, but
these approaches may differ in the stage at which the hypothesis is derived.
The difference does not dispense with the need, at some stage, for developing
an hypothesis which must be tested against data and reason.

Those of us working in geomorphology have a particular interest in the
philosophy of research, both because of the nature of our subject and the history
of its development. The aim of this portion of the geologic science is to under-
stand the forms of the earth's surface. It is not difficult to see, then, that it is
a subject which might first have been approached by classifying the observed
forms, i.e. devising categories for pigeonholing different types of hills, valleys,
scarps, rivers, and drainage patterns. From such classification, certain general-
izations were drawn an inductive approach.

A continued interest in classification, during the first third of the present
century, took the form of assigning names to features of the landscape. Streams
were designated as subsequent, superimposed, etc., and each such designation
carried with it appropriate inference about both operative processes and his-
torical sequence. Little attention was paid to the study of process, which,
looking back at the record, now appears to have led to a neglect of field studies
as the foundation of geomorphic science. As a result, the subject became one
of decreasing interest to other workers in geology. An important aspect of
this growing disinterest was that geomorphology, as practiced, seemed to lose
its inherent usefulness.

In science usefulness is measured in part by ability to forecast, i.e., to predict
relations postulated by reasoning about associations and subsequently subject
to verification by experiment or field study. With this in mind, it is apparent
that preoccupation with description could lead to decreasing usefulness because
classification and description are usually insufficient bases for extrapolation
and thus for prediction.

At mid-century there began a revitalization of geomorphology, which has
taken the form of a more detailed investigation of processes operative in land-
scape development. Study of process has been accompanied by increased use
of quantitative data and mathematical expression. The trend toward quanti-
tative study in geomorphology, in contrast with description, should not be
viewed as a basic difference in method of investigation, as mentioned earlier,
but rather as a difference in the type of problem being attacked. Both quanti-

188 LEOPOLD AND LAXGBEIX

tative and qualitative geologic research are based on the use of associations
and the concept of uniformity.

This trend parallels that in geologic research in general. Before 1930, less
than one page in a hundred in the '"Bulletin of the Geological Society of
America" contained mathematical formulation. The percentage now ap-
proaches ten in a hundred. Civil engineering shows a similar trend, but the
level has always been higher.

Coupled with the forward increase in the quantitative method in geomor-
phology there is, encouragingly enough, a greater concentration on geologic
mapping in many investigations, and in others, at least a detailed study of
stratigraphy in the field. The work of John T. Hack (1960) on geomorphology
of the central Appalachians is a model which it is hoped a growing proportion
of workers in the field will emulate. He made detailed geologic maps of local
areas which he then used as the basis for study of form and process in which
both qualitative and quantitative arguments were used. Similarly, John P.
Miller mapped extensive areas in the Sangre de Cristo Range, New Mexico,
and used these as the basis for geomorphic studies (see Miller, 1959).

Among the geologic sciences, geomorphology has, for some time in America,
been approached in a manner sufficiently different from other aspects of geology
that it may have come to be viewed as different in philosophy. We contend
that in philosophy and in method it is one with other geologic sciences. Asso-
ciation, uniformity, working hypothesis, reasoning, quantitative and qualitative
data are concepts and tools as much needed here as elsewhere in geology.

At any time the need for a set of questions, implicit or explicit, is paramount.
Over and above that, there is a time for new data and there is a time for new
theory. Progress depends on both. For several decades governmental authori-
ties had been collecting data on rivers. No one knew just how to apply this
store of information to geomorphic inquiry. No one knew what questions to
ask. Then, in 1945, Horton set forth a new hydrophysical theory of the land-
scape that was refreshingly exact in its principles in contrast with the anthropo-
morphic word pictures of William Morris Davis. The analysis of river data
began soon after. There followed a decade and a half of analysis using the data
available hi conjunction with the ideas stimulated by Horton's theory. Not
much more is likely to be gleaned from either. The time is set for new theory
and new data.

The shift in interest from description toward process and from the qualitative
toward the quantitative in geomorphology appears also to be leading toward
an important shift in viewpoint which may have far-reaching effects on the
field. New sets of associations are evolving because of the particular questions
now being asked. We think we see operating in landscape development a
principle long recognized in physics but new to geomorphic thinking a prin-
ciple of indeterminacy.

ASSOCIATION AND INDETERMINACY IX GEOMORPHOLOGY 189

By indeterminacy in the present context we refer to those situations in which
the applicable physical laws may be satisfied by a large number of combina-
tions of values of interdependent variables. As a result, a number of individual
cases will differ among themselves, although their average is reproducible in
different samples. Any individual case, then, cannot be forecast or specified
except in a statistical sense. The result of an individual case is indeterminate.

Where a large number of interacting factors are involved in a large number
of individual cases or examples, the possibilities of combination are so great
that physical laws governing forces and motions are not sufficient to determine
the outcome of these interactions in an individual case. The physical laws
may be completely fulfilled by a variety of combinations of the interrelated
factors. The remaining statements are stochastic in nature rather than physical.

These stochastic statements differ from deterministic physical laws in that
the former cany with them the idea of an irreducible uncertainty. As more is
known about the processes operating and as more is learned about the factors
involved, the range of uncertainty will decrease, but it never will be entirely
removed.

An example may be drawn from river processes. Into a given reach of river
between tributaries, a certain rate of flow of water and a certain amount of
sediment are introduced from upstream. Both change during the passage of a
given flood or through a season or a period of years. To accommodate these
various rates of discharge of water and sediment, a number of interdependent
hydraulic variables will change, including width, depth, velocity, slope, and
hydraulic roughness. A particular change in discharge and sediment may be
accommodated by several combinations of values of these dependent or adjust-
able factors.

Specifically, the physical equations which must always be satisfied are
equations of conservation, such as the conservation of mass. In the river, this
is expressed in the statement,

Q. = "&>

or discharge is the product of width, depth, and mean velocity. Another
physical equation is the relation of velocity, depth, slope, and hydraulic rough-
ness expressed by the Chezy or Manning equation. Another is the relation
between shear stress and the sediment load. In a particular case these physical
relations can be satisfied by a variety of combinations of values of the dependent
variables.

In addition to the physical laws of conservation, another kind of principle
is operating, a principle which deals with distribution of energy in time and
space and is probabilistic in form. It operates as tendencies guiding the com-
bination of the dependent factors. There is a tendency toward minimum work
or minimum rate of energy expenditure and, separately, a tendency toward

190 LEOPOLD AND LAXGBEIX

uniform distribution of energy expenditure. These are usually opposing
tendencies. These tendencies operate through processes which tend to keep an
equilibrium among the factors by restraining change.

In the river, such processes, or governors, include scour and fill, changes in
bed configuration (ripples, dunes, and antidunesj, and the Bagnold dispersive
stress on the bed. Such processes act in the same manner as the mechanical
governor on the old steam engine. Any tendency to change one factor at the
expense of another induces a resistance to that change, and so the hydraulic
factors hover around a mean or equilibrium. But at any moment in time, the
specific relations cannot be forecast except in a statistical sense.

Such governing action is well known in the process of scour and fill. If local
deposition occurs on the bed of an alluvial channel, depth tends to decrease
slightly, velocity may increase, and slope may tend to increase, the net result
of which tends to limit deposition or to induce compensatory scour. The
average relation or the most stable relation in river mechanics appears to be
one in which total energy expenditure is minimized and energy utilization is
uniformly distributed through the channel reach, a consequence of the require-
ments for a stable open system (Leopold and Langbein, 1962).

In the development of land forms there are many different processes acting
at innumerable localities. There are, in other words, a great many hills, rills,
valley's, cliffs, and other forms, and on each, a large number of variable factors
operate. Geomorphologists have considered that the variations observed
among examples of the same features are due to two principal causes: (a) slight
variations in local structure, lithology, vegetation, or other factors, and (b)
irreducible errors in measurements. We postulate a third no less important
one statistical variation resulting from the indeterminacy discussed above.
At first blush, this addition may seem trivial, obvious, or implied in the first
two causes, but philosophically it seems important. The following example
may illustrate the point.

Imagine a broad hill slope of uniform material and constant slope subjected
to the same conditions of rainfall, an ideal case not realized in nature. As-
sume that the slope, material, and precipitation were such that a large num-
ber of rills existed on the surface in the form of a shallow drainage net. Would
it be supposed that rills comparable in size and position were absolutely identi-
cal? The postulate of indeterminacy would suggest that they would be very
similar but not identical. A statistical variation would exist, with a small
standard deviation to be sure, but the lack of identity would reflect the chance
variation among various examples, even under uniform conditions.

In addition to known physical relationships there are other relations of a
stochastic nature that can be used to explain certain geomorphic forms (Lang-
bein, 1963), and they imply that variance in form is an inherent property. It
is here suggested that the same principle may have general applicability to

ASSOCIATION AND INDETERMINACY IN GEOMORPHOLOGY 191

many aspects of geologic science. The landscape, in other words, exhibits a
variability which may be expected as a result of incomplete dynamic deter-
minacy. General physical laws are necessary but not sufficient to determine
the exact shape of each land form. Some scatter of points on graphs showing
interrelations between factors is expected, although the mean or median condi-
tion is reproducible in different sets of samples.

The same set of conditions, for example the same climate, the same lithology,
and the same structure, can lead to a spectrum of different dimensions and
positions of the otherwise identical aspects, for example, the rills just mentioned.
These variations exist even though there are (a) common climatic and geologic
environment, and (b) a common set of hydraulic principles.

Hence we conclude that there remain certain unsatisfied conditions, certain
degrees of freedom (excess of unknowns over number of equations that can be
written to connect these unknowns). Implicit in this observation is the possi-
bility of applying principles of probability to an interpretation of those aspects
of the landscape subject to variance. The analysis is helped by the central-
limit theorem that a mean condition exists. The variance about the mean is a
function of the degrees of freedom.

Thus it appears that in geomorphologic systems the ability to measure may
always exceed ability to forecast or explain. The better to account for varia-
tions in land forms, it may be possible to introduce new relationships, each
deriving importance in proportion to the extent that they satisfy nature, i.e.
agree with reality in the field. These new or alternative relationships may be
stochastic rather than physically deterministic. Thus probabilistic relationships
may provide better agreement with actual conditions than the direct physical
relationships which have previously been used. The stochastic statements,
which may at times enlarge upon physical relations based on Newtonian laws
of mechanics, will differ from the latter in having an inherent variance im-
plicitly or explicitly stated. But this probabilistic or stochastic statement may
turn out to be the more important element and lead to more specific under-
standing of processes than the previous approximation which supposed exact
physical laws.

What we believe will be an example of the substitution of a probabilistic
statement for a physical one, and thus of an improvement in understanding
is in the much-studied logarithmic distribution of velocity in turbulent flow.
The approximation to field observation provided by momentum theory is
deterministic in nature, but it is well known that it contains implicitly a vari-
ance. It now begins to appear that explanation of the logarithmic velocity
distribution based on stochastic principles may be more basic in leading to
understanding and will agree as well or better with actuality than the physical
models previously used. Further, the stochastic relation will lead directly not
only to mean values but also to a statement of the variance about the mean.

192 LEOPOLD AND LAXGBEIX

Equilibrium in geomorphology, from this point of view, can be achieved
in a variety of ways and is fixed or definable for a large number of cases only
by their means. Those cases which deviate from the mean are not necessarily
in any less perfect equilibrium than other cases which coincide with the mean.
In this sense geomorphology inherently involves variance, which is an intrinsic
property of geomorphic forms.

In this light, statistical processes and statistical treatment are necessary
objects of study and tools of the science. They can be studied only quantita-
tively. This is, if need be, justification enough for the growing emphasis on a
quantitative rather than a descriptive treatment of land forms.

But justification of our tools, our methods, or our emphasis, should not
occupy attention in geomorphology*. If results are of intellectual interest, or
lend themselves to practical pre\*ision or forecasting, the science will prosper.
To this end, geomorphologists might best look to the scope of the associations
in our reasoning processes.

Any aspect of science may founder temporarily on the shoals of small ques-
tions, of details, as well as on the dead-end shallows of description. Resurgence
of activity and interest can revitalize a subject when the questions posed for
investigation are big ones, questions which, if answered, have wide applicability
or lead to broad generalization. But generalization is the broadening of asso-
ciations, the spreading of a foundation for reasoning. The big question is one
the answer to which might open up new or enlarged areas of inference or asso-
ciation.

The measure of a research man is the kind of question he poses. So, also,
the vitality of a branch of science is a reflection of the magnitude or importance
of the questions on which its students are applying their effort. Geomorphology
is an example of a field of inquiry rejuvenated not so much by new methods as
by recognition of the great and interesting questions that confront the geologist.

REFERENCES CITED

HACK, J. T., 1960, Interpretation of erosional topography in humid temperate regions:
Am. J. Sci., vol. 258-A (Bradley Volume), pp. 80-97.

HORTON, R. ., 1945, Erosional development of streams and their drainage basins,
hydrophysical approach to quantitative morphology: Geol. Soc. Am., ., vol. 56,
pp. 275-370.

LANGBEIN, W. B., 1963, A theory of river channel adjustment: Am. Soc. Civil Engrs.,
Tr., (in press).

LEOPOLD, L. B., and LANGBEEN, W. B., 1962, The concept of entropy in landscape
evolution: U.S. Geol. Survey, Prof. Paper 500-A, 20 pp.

MILLER, J. P., 1959, Geomorphology in North America: Przeglad Geograficzny, War-
saw, vol. 31, no. 3-4, pp. 567-587.

FREDERICK BETZ, JR.

Geologic Communication

Geology may be viewed as a body of knowledge that grows by additions of
observations, verifications, and interpretations. The significant point in all
identification of knowledge is that it has no tangibility until it has been ex-
pressed in terms which are understandable to at least one individual beyond
the discoverer. Thus the vitality of geology its data, hypotheses, principles,
and methods of investigation is nourished by various modes of communica-
tion, which are understood first by those who devote themselves to this science
and second by others to whom geologic information is conveyed.

While communication must remain abreast of investigation, it should not
become an end in itself. The mass of scientific information has already become
greater than any individual can absorb. Soon the scientist will not even be
able to deal effectively with the abstracts published as guides to the literature
of his special field. To cope with this problem, a "science of communication"
has developed, but this separation of communication from subject matter
brings with it the threat that the servant may come to dominate the master.

Perhaps the best way for scientists to attack the "information problem" is
to become more expert with the tools of communication, which we often use
indiscriminately and badly. In the following discussion, the example of geologic
communication is considered from several points of view. Among the topics
to be considered are the nature of geologic data, the characteristics of the tools
used to communicate geologic information, and the influence of communica-
tion upon the body of geologic thought.

Data of Geology and Tools of Communication

W. G. Krumbein (1962) classifies the data of geology as observational and
experimental. The first type consists of (a) "qualitative observations or state-
ments regarding natural objects or events" and (b) "numerical measurements
on those natural objects or events." They are obtained in the field or "measure-
ment" laboratory. The second represents "quantitative measurement data
arising under specified and controlled conditions in an experimental labora-
tory." Krumbein asserts that, although geology is basically a qualitative
science and geologists are most concerned with observational data, quantifica-
tion is advancing steadily in some subfields.

193

194 FREDERICK BETZ, JR.

The tools used to record and communicate geologic data are language,
mathematical and abstract symbols, and graphic representations. They are
used individually, but often appear in combination, partly because they are
needed to supplement or reinforce one another in providing the most effective
communication of different kinds of data, and partly because the capacity of
each tool is not fully exploited. There may also be a substitution of tools
in different stages of recording and communicating data or interpreting
them. The combined use and substitution of tools is illustrated by the geologic
map.

Since the beginnings of geology-, a major aim of its investigations has been
the mapping of geologic objects, mainly units of rock. Krumbein notes that
the measurements of the location, distribution, and orientation of units of rock
are examples of the long use of numerical data in geology*. However, the rocks
are defined by physical and chemical properties, not by these "observational"
measurements. Some of the properties are examined quantitatively, but others
are not or cannot be. The rock type is, therefore, defined by mixed quantitative
and qualitative data. Furthermore, the unit of rock that is mapped is estab-
lished on a qualitative interpretation or judgment of unity, represented either
by homogeneity or some peculiar heterogeneity of components. The judgment
is often based on the relative age assigned to the rocks by use of a variety of
procedures, both qualitative and quantitative. If users of the map or other
observers in the field accept the judgment of the mapper, it is credited with
being objective. \Vhere there is disagreement, the same area may be mapped
more than once with different results, one of which is presumed by each mapper
to be correct. The desire to promote quantification and eliminate subjective
determinations and evaluations in geologic mapping is understandable, but,
as Krumbein warns, quantification raises a problem that pervades all of geology.
It "involves making a distinction between those parts of the science that can
best be treated wholly on a quantitative basis and those that may actually be
weakened by overquantification." The danger lies in attributing regularities
to geologic objects, which, to an unknown degree, do not exist or cannot be
proved.

In the geologic map, some of the recorded measurements of dimension and
orientation are expressed graphically, but others are superposed directly on
the map sheet in the form of numbers. The quantitative reliability of the
presentation is enhanced by the use of a topographic base, which allows an
immediate visual recognition of the relationship between the rocks and the
configuration of the surface.

Data on lithology and relative age are expressed by colors, patterns, and
symbols. On large-scale maps, graphic presentation of these data need not be
a difficult exercise in communication. In maps of medium to small scales, the
quantity of data given must be smaller or the detail simplified. Simplification

GEOLOGIC COMMUNICATION 195

may be preferred because it permits retention of both categories of data. At
still smaller scales, the alternative of eliminating one category is often unavoid-
able. Since the geologic map is primarily an instrument of historical investiga-
tion, the data on lithology are sacrificed so as to retain information on relative
age.

When the data are not incorporated in the map, they are usually given in
texts and tables. When printed on the map sheet, this information can be
regarded as the explanation of the map. Frequently, however, the information
is contained in a separate pamphlet or book, where it may be combined with
material not direcdy related to the map. Then, instead of serving as an accom-
paniment to the map, the written material is the main product of the mapping
investigation, and the map is secondary. At small scales, which are used for
compilations of data covering large regions, or for indices to maps of larger
scales, all details must be provided outside the map.

As constituted, then, the map is not a self-contained communication. The
information necessary to make it completely useful must be obtained elsewhere
or by interpretation based on the user's knowledge and experience. For the
nongeologist, the cryptic information on relative age and lithology has little
meaning or value.

The question raised by the geologic map is whether it represents the best
and most complete use of the cartographic medium for the purpose. The im-
pression created by the geologic map is that the amount and nature of informa-
tion exceed the capacity of the graphic tool. It might be presumed that the
map, because of its fundamental place in geologic science, has been carefully
and steadily developed, and that the combined use of the graphic tool and other
tools was a refined solution of a problem in communication. The evidence is
to the contrary. The map is essentially what it was in its earliest examples,
a geographic directory of geologic objects only superficially characterized by
graphic devices.

That there is an opportunity for fuller use of the cartographic medium is
demonstrated by various types of maps. In geology, progress toward better
direct communication of data through maps has been made where a practical
purpose is involved. In this case, there is usually a need to make the map
usable by and useful to nongeologists. The requirement has forced geologists
producing such maps to become aware of the resources of the cartographic
medium. In the case of the geologic map, two types of investigations might be
undertaken to improve and develop it. One would be to determine whether
the map should have additional functions of communication, such as synthe-
sizing the relationships between the objects mapped and placing the geologic
data in a more distinct environmental perspective. The other would be to
analyze the potential of the cartographic medium to determine whether in-
formation now in texts and tables could be expressed graphically.

196 FREDERICK BETZ, JR.

The reliance on language for communication of geologic knowledge is evi-
dent, but as experimental and quantitative procedures become more common,
numbers and other systems of abstract symbols will make inroads on the dom-
inance of language.

The advantages of symbolization as a tool of communication are familiar,
but may be reviewed briefly here. Baulig (1956) points to mathematical
symbolization, especially algebraic, as the most perfect scientific language
because it is rigorously precise, unequivocal in meaning, unexcelled in ease of
handling, and universal in use. Symbolization also has the obvious advantage
of relative economy in use of space. Baulig notes that mathematical symboliza-
tion is suited best to expressing relationships, particularly abstract ones. The
physical sciences are in a position to avail themselves of this tool, but the natural
sciences are still in the process of describing and classifying objects and are not
ready to employ symbols exclusively. They must rely on the word to give form
and reality to their data and the inferences derived from them. Furthermore,
so long as substantial parts of geology are treated better qualitatively or elude
quantitative investigation, symbols will not replace language as the primary-
tool.

The advantages of mathematical symbols are the drawbacks of verbal
language. Nevertheless, scientific verbal languages possess a superiority over
systems of symbols because they are not limited to their symbols, or terminology*,
but can make use of words that everyone knows in common languages.

The deficiencies of verbal language can be overcome by accurate and skillful
use, which should be desired by the communicator and expected by his audi-
ence. Inaccurate and inept use of numbers is censured and casts doubt on the
validity of the data being presented and the investigations underlying them.
The user of language should search for a similarly strict standard of accepta-
bility*. As a group, geologists, who are dependent on language, do not give
it such support. Often those geologists and other scientists who use language
well are looked upon with suspicion, as though they were guilty of concealing
faulty procedure and inadequate data by clever manipulation of words. An
equally negative attitude common among geologists ascribes to all an innate
inability to use language correctly and effectively. The prevalence of poor
writing is, however, not to be excused in a science that requires verbal com-
munication.

Criticism of Geologic Writing

The published criticism of geologic writing focuses on poor grammar, bad
habits of style, and abuse of terminology.

The first two charges have been discussed lately by Weber (1957) in respect
to German geologic writing. This critic attributes the decline in quality of

GEOLOGIC COMMUNICATION 197

writing in recent decades to a lack of training of geologists in classical languages.
He sees the trend continuing as the poor writing of today becomes the model
for tomorrow's authors. To substantiate his claim Weber refers to errors in
syntax and spelling, use of dialectal words, and diffuseness. Earlier, Cloos
(1933) had protested against the wordiness of German geologic literature,
illustrating his complaint by examples that he rewrote in shorter, clearer form.
Geologic literature in other languages seems similarly afflicted.

These criticisms can be debated, because grammar and style are themselves
not clearly defined. Usage has replaced grammar in some languages, particu-
larly English, as the standard of correctness and acceptability. Bergen Evans
(1962) says: "There is no simple rule about English that does not have so many
exceptions that it would be folly to rely on it." The obligation of the writer
in English has shifted from strict observance of grammatical rules to an aware-
ness of accepted ways of saying things. This obligation rests also on editors,
who have the power to destroy good writing by application of bad rules.

Referring to scientific and technical translating, which shares the problems
of original writing, Holmstrom (1951) says that the ability to express thought
"involves skill in exploiting the possibilities of interplay between words through
syntax and accidence, emphasis and cadence, metaphor and simile, sentence
and paragraph structure and the indefinable thing called 'style.'" The diffi-
culty of acquiring this skill is stated by E. B. White, who says 1 that "There is
no satisfactory explanation of style, no infallible guide to good writing." Evans
advises:

In order to exercise good taste in English one must know the full spread of
what is allowable, the great variety of forms that are being used by the best
speakers and writers. And one learns this not by mastering rules but by
paying careful attention to details in the speech or writing of those whose
English seems attractive or compelling.

The advice is given without reference to subject matter or intended audience.
The criteria of good style are as relevant to technical writing as to literary
writing and perhaps should be applied more stringently to the former, if one
accepts the views of Darlington (1953):

The theory that scientific discovery is impersonal, or as it is called, objec-
tive, has had several evil consequences . . . One should write, one is told, in
the third person, in the passive voice, without betraying conviction or em-
phasis, without allusion to any concrete or everyday object but with the
feeblest indifference and the greatest abstraction. This practice has proved
to be so readily acquired that it has now, for a whole generation, been
debauching the literary language of the world. The result has been that

1 See Strunk and White in the References.

198 FREDERICK BETZ, JR.

science, instead of being a source of strength and honesty, is fast robbing
the common speech of these very qualities. For the style itself is neither
strong nor honest.

Several decades ago, Sir James Barrie was quoted by T. A. Rickard (1931)
as having said that "the man of science appears to be the only man who has
something to say, just now and the only man who does not know how to say
it." Rickard preferred to think that this comment had been a friendly jibe at
the pure scientists of the nineteenth century, but was an unpleasant truth when
applied to practitioners of "the avowedly utilitarian branches of science" of the
present day. A more significant comparison would be made between present-
day writing in pure science and applied science. In the case of geologic writing,
as with geologic maps, clarity of communication seems to be more and more an
attribute of applied science, because it is a necessity. The unfavorable criticism
should be aimed at geologists who feel no compulsion to write well, not at
branches of their science. Barrie's premise that the scientific writer has some-
thing to say is put to a serious test by many writers. As a consolation they may
accept White's opinion that there is "no assurance that a person who thinks
clearly will be able to write clearly." 2 One may also note that there is no assur-
ance that a person who does not write clearly is able to think clearly.

To combat the weaknesses in grammar and style of geologic writing, Weber
(1957) suggests two measures. One is to encourage systematic scrutiny of
published literature for offenses, which should be publicly cited. The other is
to encourage authors to assume personal responsibility for correcting and im-
proving their writing before submitting manuscripts for publication. The first
proposal relies on the efficacy of threats to improve behavior. It would cer-
tainly meet with widespread opposition, even from those not found guilty.
The second is the more agreeable proposal, but it rests on the dubious assump-
tion that geologists as a group are better critics of writing than they are writers.

An improvement in geologic writing may result from the strain being placed
on publication facilities by the constantly increasing number of manuscripts
submitted. Among the ways of relieving this strain, several that have been
introduced in scientific and technical journals strike at prolixity. The objective
is attained by setting maximum limits on the length of manuscripts that will
be published or by offering incentives to authors to write as briefly as possible.
A promise of more rapid publication is the most persuasive argument. In
acceding to self-discipline, more authors may learn to appreciate the weight
and value of words.

The third criticism of geologic writing concerns the abuses of terminology.
C. W. Washburne (1943) describes them as the use of incorrect terms, contra-

2 See Strunk and White in the References.

GEOLOGIC COMMUNICATION 199

dictory terms, pleonasms, and fancy words, and the misuse of valid terms. A
distinction can be made between the more and the less serious abuses. Pleo-
nasms are often idiomatic and have a claim to acceptability, especially in the
spoken language. The use of fancy words, including foreign-language equiva-
lents and obsolete terms, can be altogether appropriate in places. White
treats these offenses under the heading of style rather than correct or acceptable
usage. The inexcusable abuse is the use of a valid term that is irrelevant in
the context.

Watznauer (1956) sees a twofold purpose in terminology. To the individual,
terminology is the means of possessing knowledge in condensed form; to the
group, terminology provides the verbal codes that simplify communication.
Terminology should be a distinct aid to scientists, but has become increasingly
a barrier to understanding. The barrier is heightened by the proliferation of
specialized vocabularies. When communication based on the use of terminology
breaks down, the special language becomes a jargon.

Challinor (1961) attempts to clarify the problem by differentiating between
jargon and acceptable terminology. He calls jargon "Twittering, chattering.
Terminology invented or used where ordinary words would do as well." On
the other hand, he says "Technical terminology, geological terminology, how-
ever uncouth the terms, is not jargon if it fixes meaning shortly and concisely."
Regrettably, his attempt fails, for neither gracefulness nor succinctness pre-
vents terminology from being jargon, even when the terms look like ordinary
words.

There is also the problem of communication with scientists in other fields
and with the general public, which terminology tends to complicate. A drastic
view of the barrier that exists here is shown in the following statement from
a geologic dictionary for general users by Himus (1954): "It is a common
accusation that geologists in common with other scientists are guilty of in-
venting and using a barbarous and repellent jargon which is incomprehensible
to the man in the street and may even be adopted as a protection against
criticism."

Such criticism of geologic writing underscores the need for skill and accuracy
in the use of language. Leniency may be permissible in judging skill, but accur-
acy, particularly in the use of terminology, is essential. There must be a com-
monly held belief in the integrity and inviolability of terminology to which both
the user and the creator of terms subscribe. Acceptance of this belief would
prevent the user from using terms without knowing their meanings, and the
creator from introducing terms for which there was no need. The abuses by
the user can be overcome by the conscientious use of dictionaries and other
types of wordbooks. The creator has the added obligation of knowing the
motivation for terms, the sources of terms, and the terminology of his subject
in different languages.

200 FREDERICK BETZ, JR.

Sources of Geologic Terminology and Implications of Its Growth

In his work on the composition of scientific words, Roland Brown says that
these words "originate in three ways: (1; adoption directly with appropriate
modifications in spelling from Greek, Latin, and other languages; (2) composi-
tion by compounding and affixation; and (3) outright or arbitrary creation
without use of evident, antecedent root or stem material."

It is not always easy to determine which of these actions accounts for the
origin of a particular term. Some dictionaries explain the etymology of the
words they contain, but this information does not tell whether the first user of
a term borrowed or derived it from an existing word or invented it. In fact,
many terms, especially the older ones, entered the vocabulary without a known
formal introduction. The source shown in dictionaries is more often the author
of an accepted current definition than the first user or originator of the term.
For the user of terminology, the accepted meanings are, of course, far more
important than the history of usage, and it would not serve his purpose to have
this additional historical information.

F. A. Burt (1949) has examined the origin of geologic terms specifically and
found that the commonest source is the vernacular. He cites such terms as
joint, graben, dune, lava, and moraine as examples of ordinary words that have
entered the geologic vocabulary, but with more restricted meanings than they
had originally. There is an almost endless list of common words with geologic
meanings. To add only a few, ash, axis, basin, bed, bomb, boss, groove, group, habit,
head, heave, and incompetent can be mentioned. Burt also refers to terms that
come from particular environments, such as barchan, hamada, and nunatak.
Baulig (1956) remarks that the terminology of geomorphology is filled with
such words, which come into use as localisms and, in some cases, acquire wider
use.

Besides common words, proper nouns are adopted or adapted for use in
the geologic vocabulary. Every geologist is familiar with the many stratigraphic
and rock names that are derived from specific geographic designations and the
fossil names drawn from surnames. Burt cites the following examples of terms
of this origin: in geomorphology, meander and monadnock; in paleontology,
Beltina; in mineralogy, labradorite; in petrology, syenite; and in ground-water
geology, artesian.

It is evident that many cases of adoption or adaptation of ordinary words
and proper names involve borrowing from a foreign language. Decke, fjord,
graben, horst, Jdippe, loess, and nappe are examples of terms, used in geologic
vocabularies generally, which have been borrowed from vocabularies of par-
ticular languages.

Terms should be borrowed from foreign languages primarily to fill gaps in
terminology. A second reason for his kind of borrowing is to take advantage
of terms that seem especially suitable to subject-matter specialists. The latter

GEOLOGIC COMMUNICATION 201

reason is less definite than the former, which can be determined by the inspec-
tion of vocabularies in different languages. Gaps are far more frequent, not
only in scientific terminology but also in ordinary words, than is usually
realized.

To illustrate the gaps in geologic terminology, a sample has been taken from
the four-language geologic dictionary edited by Rutten. The section on "tec-
tonic geology" contains 653 terms (not including synonyms), of which only
393 are represented in all four of the languages, 141 in three, 105 in two, and
14 in one. There are 644 terms in Dutch, 561 in German, and 506 in both
English and French. Dutch appears to have the richest terminology of this
subject, but the relative scarcity of gaps is undoubtedly related to the fact that
Dutch was the primary language of the compilers. Furthermore, the dictionary
project afforded them an opportunity to discover the gaps in the Dutch ter-
minology, which, according to two of the compilers, were closed by coining
terms.

The absence of equivalents in two of the four languages for any term is less
significant than a gap in one or three. Where at least two languages contain
equivalents, the sense of the knowledge they express is seen to have gained
identification beyond its source. A term that has no equivalents in three of the
languages may represent a highly restricted identification, although it may
also stand for a distinct advance of knowledge in one language area over the
other areas. The absence of a term in only one of the languages suggests that
the subject has, for some reason, not been pursued actively in that language
area. This interpretation is supported by the prevalence of gaps in sequences
of related terms. On the other hand, it may indicate that foreign-language
terms are being used regularly to discuss the subject.

In the sample, English and French show the same number of gaps, but
English is the only missing language in 83 cases, whereas French alone is absent
in 42. Based on a study of the terms of geomorphology, Baulig finds that
foreign-language terms are used most in English and successively less in German
and French. He relates the prevalence of borrowed technical terms in English
to the long-time general receptiveness of English-language speakers to words
of foreign origin. German, according to Baulig, is more likely to contain trans-
lations of foreign-language terms or modifications to a German form. French,
he finds, resists German terms because they seem difficult for French speakers
to assimilate and pronounce, but is more tolerant of words from other Romance
languages. Thus, the indicated gaps in French terminology are more likely to
be real gaps than those in English.

Baulig comments on the dangers of borrowing without adequately under-
standing the source language. Too often, adoptions result in distortions of
meaning and incorrect usage. The tendency to borrow freely makes these
abuses all the more noticeable in English writing and speech.

202 FREDERICK BETZ, JR.

Borrowing occurs also between different sciences, particularly to serve the
needs of specialists working in border areas. The rapidly developing inter-
disciplinary fields, such as geophysics and geochemistry, have vocabularies
that represent mingling, rather than mere borrowing, of terms from the parent
fields. As terms become common to more than one field, they tend to fall
within the scope of the general language and appear in standard language
dictionaries. The commonly accepted definitions of an increasing number of
scientific terms are found today in these dictionaries, which are cited as the
authorities for definitions shown in modern technical dictionaries.

Adoption or adaptation of words from any source often involves change in
meaning. The scientific usage is subject to further change, with the result that
many terms have several different scientific meanings, not just in succession
but simultaneously. Watznauer regards change from original meanings as
normal deviation caused by the advance of knowledge. A typical example of
the history of origin and changes in usage has been traced by Tourtelot, using
the word sfale . He concludes that the present usage of this term hi two technical
meanings is permissible and justifiable. It would be futile to insist on a single
meaning for this kind of term. The user cannot be expected to define his usage
of each term with different meanings each time he uses it. The usage, whether
technical or vernacular, should be apparent to the reader from the context.

Invention of terms should become necessary only when no terms fulfilling
the purpose are available in any language. The need grows as the knowledge
of the field increases. The discovery* of hitherto unidentified objects demands
the creation of new names. There is less frequent need for new common words,
since they stand for classes of objects, generalizations, and concepts, which are
not newly recognized as often as data are.

Whatever the mode of origin, need is the basic justification for terminology.
The growth of terminology in modern time suggests that need does not always
motivate the introduction of new terms. Watznauer has examined this growth
critically and concluded that the purpose of terminology is being destroyed by
uncontrolled addition of specialized terms. He argues that a new term should
represent an addition to knowledge and that too many terms are unnecessary
because they merely reexpress what has been known before. He observes that
terms are being introduced without clear-cut definitions and therefore do not
serve as a point of reference for the state of knowledge at a given time.

Watznauer (1956) sees the individual scientist unable to cope with the many
terms applying to special subfields and abandoning the effort to keep abreast
of knowledge in specialties other than his own. He finds terminology- threaten-
ing to become the master of science and artificially promoting specialization
that tends to disunite fields. This argument oversimplifies the explanation of
real or apparent fragmentation of sciences. The actual increase in the store of
knowledge is still a compelling factor in limiting the scope of individual interest

GEOLOGIC COMMUNICATION 203

in and capacity for comprehending a science. In the case of geology, specialized
vocabularies do not cause the drift of subfields toward affiliation with other
sciences, but once the movement has begun they serve to reinforce it.

Monolingual Dictionaries

Dictionaries and other types of wordbooks are a measure of the amount
and kind of attention given the subject of terminology. Bibliographic and sta-
tistical records of all types of monolingual and multilingual dictionaries through-
out the world, published by UNESCO (1961 see also Wiister, 1955, 1959) show
geologic dictionaries to be few in number today; and there is no evidence that
they were more common in the past.

Dictionaries can be classified as general or specialized by reference to their
coverage of fields and degree of completeness within the fields covered. In a
strict interpretation, a general dictionary is one that covers the entire termin-
ology of a field at a given time. An all-inclusive geologic dictionary may have
been compiled in the past, but none appears to exist now. By design, the
modern general geologic dictionary is one with broad coverage of the termi-
nology, but with intentional omissions of some subject areas or categories of
terms.

Specialized dictionaries do not have to conform to any standard of coverage
or completeness. These dictionaries may be separate publications of consider-
able size with complete coverage of a stated range of subject matter, but often
they are glossaries and vocabularies appended to geologic books, papers, and
map explanations.

In English, four general geologic dictionaries published in the nineteenth
century by Roberts in 1839, by Humble in 1840, by Page in 1865, and by
Oldham in 1 879 are cited in modern dictionaries as predecessors. The modern
dictionaries, which have been examined for this discussion, are those prepared
by Rice (1940), Himus (1954), Stokes and Varnes (1955), the AGI Glossary
Project (1957; second edition containing a supplement, 1960; abridged ver-
sion, 1962), and ChaUinor (1961).

The Rice and AGI dictionaries are comprehensive works intended primarily
for professional geologists. The Rice dictionary (1940), with an estimated total
of 8000 terms, was the largest English-language geologic dictionary at the time
of its publication. With 14,000 terms in the original edition and about 4000
more in a supplement, the AGI dictionary is the closest approach to an all-
inclusive geologic dictionary available today. Some of those compiled in the
nineteenth century may have covered the smaller body of terminology of that
time as completely. For example, Humble (1840) defined about 3000 terms in
his dictionary published in 1840 and added 300 in the second edition in 1843.

204 FREDERICK BETZ, JR.

The difference in size of the Rice and AGI dictionaries is partly an indicator
of the growth of terminology* in a period of less than 20 years. It is also an
indicator of the vastly greater resources of manpower applied to the AGI
Project. After about four years of planning by representatives of numerous
organizations, topical committees, in which more than 80 individuals partici-
pated, spent three years and four months in selecting terms and providing
definitions. The Rice dictionary was compiled by one person, not a profes-
sional geologist, without organized assistance, over a period of many years.

As comprehensive works, both the Rice and AGI dictionaries necessarily
contain a broad coverage of geologic fields. The compilers of the AGI diction-
ary considered 30 fields of geology and related sciences and included all but a
few in the final compilation. The terminology of each field covered is complete
except in a few cases, for reasons that are stated in the preface to the dictionary.
Limited coverage was justified mainly by the existence of accessible specialized
dictionaries or other reference works on terms and names of a subject area.
In the Rice dictionary, the only announced exclusion was stratigraphic and
paleontologic nomenclature. Actually, it contains a great many stratigraphic
names.

The Himus dictionary (1954) and the abridged version of the AGI dictionary
are classed as selective general dictionaries. Both were published in series of
pocket-sized reference books intended for the general reader. The Himus
dictionary is the smaller volume, containing about 1800 terms and names.
The AGI abridged dictionary, with 7000 entries, or about half the contents of
the second edition of the original dictionary, is, despite its format, a compre-
hensive work. It establishes a new scale for selective geologic dictionaries.
Two decades ago the most complete dictionary was less than 10 percent larger
than this abridged work of today. The largest selective dictionary published
previously contained less than half as many terms.

Stokes and Varnes (1955) selected 2800 terms covering the field of geology
broadly, which they considered to be of concern to civil engineers and related
specialists. The specific omissions are stratigraphic names, most rock names,
and most general and elementary terms of mineralogy and petrology. Whether
the intended audience has no use for the information that is omitted is ques-
tionable. What Stokes and Varnes have accomplished is to put the problem
of abridgment in focus. They have distinguished information that is properly
recorded in an abridged general dictionary from that which belongs in un-
abridged general and specialized dictionaries.

Challinor (1961) explains his purpose in compiling a geologic dictionary
with this statement:

There appears to be room among works on geology for one that will probe
the subject by examining the meaning and usage of names and terms that

GEOLOGIC COMMUNICATION 205

stand for the more significant things, facts, and concepts of the science. This
small book is an essay towards a critical and historical review of a selected
ABC of the subject.

The format of a dictionary is employed by Challinor to present comments
on geologic subject matter. Terminology is treated as a manifestation of the
state of geologic knowledge and the degree of maturity of the science at different
times in its history. Secondarily, Challinor supplies a study on the usage,
source, and etymology of geologic words. There is no pretense of complete
coverage in this selection of about 1250 terms and 300 names belonging to
subfields of geology, to which 250 general geologic words, designation of tools
and concepts, and terms borrowed from physics and chemistry are added.

In addition to the five dictionaries mentioned, two other modern dictionaries
contain a broad representation of geologic terminology, although they are not
specifically geologic. These two are the dictionary of mining- and mineral-
industry terms by Fay (1920), and the geographic dictionary prepared by a
committee (L. D. Stamp, Chairman) of the British Association for the Advance-
ment of Science (BAAS), which was published in 1961.

The Fay volume (1920) contains 20,000 terms, or 8000 more than appear
in the first edition of the AGI dictionary. In this compilation, Fay had six
active collaborators and twelve reviewers of definitions. About 4000 of the
terms can be considered geologic. Thus the Fay dictionary was also the first
major geologic dictionary published in the United States. Only paleontologic
terms are specifically omitted, but, in keeping with the purpose of the work,
the majority of the geologic terms are mineral and rock names.

The BAAS dictionary (1961) may be regarded as the geographic counterpart
of the AGI dictionary, since it, too, is the result of an intensive examination
of terminology for the purpose of producing a comprehensive general diction-
ary. The BAAS project took even more time than the AGI project. In a first
stage, an attempt to carry out directions given by the BAAS to a dictionary
committee ended in a virtual suspension of effort after three years. The
project was resumed by a reorganized committee of twelve members, supported
ultimately by fifty-eight collaborators, many of them outside Britain. A re-
search officer was appointed to prepare an initial list of terms and definitions.
This task was completed in one year. In the following six years, Stamp, who
was also the editor of the dictionary, quadrupled the list of terms and, with the
assistance of the other committee members and the collaborators, reviewed
definitions and prepared comments on them.

With an estimated total of only 3200 terms, the BAAS dictionary still may
be classified as both an unabridged and an abridged dictionary. The size does
not indicate selectivity in subject-matter coverage so much as intentional
limitation to "terms in current geographical literature written hi English."

206 FREDERICK BETZ, JR

The criterion of currency is not applied in any of the other dictionaries being
considered here.

The BAAS dictionary contains many terms used in geography that are also
identified with the vocabularies of geology*, climatology-, meteorology, soil
science, and ecology*. In respect to geologic terminology, Stamp notes the
following omissions: mineral names; rock names except for some "commonly
misused or misunderstood ^names,' and names applied to broad groups' 5 ; all
but a few stratigraphic names (but more than 200 are listed in an appendix to
the dictionary;; and all but some important tectonic terms. These omissions,
primarily of nomenclature, do not indicate a narrowness of interest in geologic
subject matter among physical geographers. Geologists might presume that
the majority of terms in the dictionary were taken from the field of geomorphol-
ogy. Actually, geomorphic terms amount to less than 10 percent of the total
of geologic terms. The coverage of geologic terms is so broad that, according
to a review in '"The Geochemical Xews," 3 the dictionary contains many terms
of interest "to geologists, mineralogists, petrologists, and even geochemists. 59

The fundamental significance of the widespread borrowing and sharing of
terminology and subject matter among sciences is that they no longer have
completely separate areas of interest. Indeed the boundaries between the major
fields are being erased rapidly. The question can be asked: What will be the
identifying marks of the individual fields in the future? An answer is given for
geology by C. \V. Wright (1959), who characterized this field as a group of
sciences unified by a special outlook, which he describes as a sense of time and
process. Wright mentions only chemistry, biology, geography, and meteorology
as sciences joined for a special purpose in geology, but still other sciences are
familiar in their geologic orientations. If Wright's interpretation is accepted,
then one can ask whether geology was ever anything but an outlook on prob-
lems dealt with from other standpoints by other disciplines. Today, the identi-
fication of such fields as geophysics, geochemistry*, and geohydrology is viewed
widely as evidence of the fragmentation of the science of geology. On the con-
trary, these subfields are interdisciplinary* links between geology and other
sciences that have always existed but were not recognized formerly. In the
process of identifying these links, the concept of geology* as a completely sep-
arate discipline is becoming obsolete. lake other sciences that have already
abandoned an imaginary autonomy, geology will become identified as one
important viewpoint influencing investigations in a group of interdisciplinary
fields. In them the geologic influence may be dominant, but from time to time
and place to place it will undoubtedly be supplanted by influences of other
participating disciplines. It is unrealistic to classify* these fields as branches of
any one of the participating disciplines, even if they were first recognized within

* See H., in References.

GEOLOGIC COMMUNICATION 207

single disciplines. Thus geophysics is as much a branch of geology- as it is of
physics, geochemistry is not the exclusive property of geology- or chemistry, and
it is proper for both geographers and geologists to be geomorphologists.

The replacement of distinct sciences by interdisciplines has been foreshad-
owed by the terminology gathered in modern technical dictionaries. For prac-
tical reasons, comprehensive general dictionaries are forced to omit whole
vocabularies of existing subfields of the science and interdisciplines in which it
participates. Basically the compilers are faced with the problem of defining
the limits of the field and are finding the solution increasingly elusive. The
general dictionary must, therefore, abandon comprehensiveness, and the burden
of recording the complete terminology must be distributed among specialized
dictionaries.

A stated function of the BAAS dictionary- is to serve those whose native
tongue is not English, because English is becoming "increasingly the normal
medium for scientific writing and publication." This conclusion was con-
firmed by the revisers of the Royal Netherlands Geological and Mining Society's
multilingual geologic dictionary. As a Dutch publication, in the original
edition in 1929 Dutch was the language of reference. In the revision published
in 1959, English occupied this position because the compilers found that it
was becoming dominant in geologic literature.

These dictionary compilers refer to the dominant medium as English with-
out describing this "English" that is used for scientific communication. In
respect to terminology, as was indicated in the earlier discussion of sources of
terms, a significant part consists of foreign-language terms adopted with little
or no change by English-language speakers. Thus the BAAS dictionary-
contains about 900 terms, more than one-fourth of the total, from 31 foreign
languages or language groups. Presumably these terms have been assimilated
into English and are now integral parts of the communication medium called
English. In reality, an international terminology is evolving.

The natives of foreign-language areas use the so-called English terminology
in a framework that is also referred to as English. On examination, this lan-
guage is found to be a geologic lingua franca composed largely of English words,
used frequently in ways unfamiliar to English-language speakers, and of
English-like words that seem to be invented more or less spontaneously by the
foreign-language speakers. In grammar and style it often bears greater affinity
to some foreign language than to English. Foreign-born scientists carry this
lingua franca into English-language areas, where, unfortunately, too many
continue to use it. They complicate the problem of scientific communication
by adding negligence to the already prevalent indifference toward, and defiance
of, good use of language.

The difficulties created by the use of foreign-language terms and the faulty
use of English by foreigners are augmented by the differences between North

208 FREDERICK BETZ, JR.

American and British English. The BAAS dictionary takes into account
these differences in the use of terminology-. They occur in both terms and
usage. The more troublesome are the differences in usage, which are apt to
be overlooked by English-language speakers. The extent to which British
terminology* differs from American is demonstrated by the Himus geologic
dictionary. Roughly 400 of the estimated total of 1800 terms are British terms
and names, which are not used regularly, if at all, by Americans and are no
more meaningful to them than foreign-language words.

In group projects it is difficult to obtain agreement on definitions. The
BAAS instructed its first dictionary- committee to prepare "a glossary of geo-
graphical terms with agreed definitions in English including references to
origin and previous usage and to current use and misuse." The reorganized
committee evaluated the charge and concluded that the effort to prepare
definitions that would meet with complete approval of the members would
have to be abandoned. They discovered further that while original definitions
can be found for recent terms, many older terms have gained acceptance
without benefit of a specifically recorded first definition. The committee chose
to accept from a standard source, usually the Oxford English Dictionary, at
least one clear meaning and to obtain variant usages from the technical lit-
erature.

The AGI dictionary offers a combination of different types of explanations.
Presumably the topical committees adhered to the aim of finding the original
or another authoritative definition before writing a new one. The agreement
on either accepting an existing definition or preparing a new one was a responsi-
bility of each committee. These committees ranged in size from one member
to twelve, with an average of about three. There is visible evidence of a lack
of uniformity in handling definitions, which, like the variations in coverage of
topics, is probably the result of wide delegation of responsibility. That such
irregularities are not desirable was recognized by the compilers, but those
that occurred were not corrected because too much time was believed to have
been spent on the project to permit further delay in publication. Instead, an
immediate review of the dictionary for omissions and errors, in which the readers
were invited to participate, was announced. It yielded about 4000 additional
and redefined terms. The revision of the dictionary seems to prove that the
number of experts who can be expected to reach agreement on the proper
selection and definition of terms in their own special vocabularies is very low. 4

4 The following statement appears on p. iii of the AGI dictionary: "Selection of
definitions used was made by the members of the sub-committees listed below. Un-
accredited (sic!) definitions were written by these committees or individuals." As
printed, the second sentence says that definitions not ascribed to a source are not
approved definitions. This expression of no confidence in the committees and their
members can be interpreted by each reader to his own satisfaction.

GEOLOGIC COMMUNICATION 209

Interlingual Dictionaries

The basic purpose of interlingual dictionaries is to coordinate terminology'
in different languages as an aid in reading and translating foreign-language
writings. The function is to provide equivalents, not definitions. Such diction-
aries are, as Holmstrom (1951) points out, "apt to foster the illusion that each
word in a language necessarily has its exact and self-sufficient counterpart in
other languages." The presence of gaps, or absence of equivalents, has been
illustrated earlier in this discussion by the example of geologic terminology in
four languages. Where equivalents do not exist, the reader or translator must
be prepared to visualize or express the term in some other way, usually by a
phrase. He will also have the opportunity of reducing a statement in one
language to a word in another.

A technical word or expression is surrounded by other words that create the
context. The real unit of equivalence between languages is, therefore, more
often a phrase or a sentence than a word. Holmstrom (1951) states that

the first principle of good translating is to translate not words but ideas
organically. The effect of reading a sentence in one language ought to be to
generate in the translator's mind an articulated pattern of ideas, so that he
may then forget the wording in which they were originally expressed and
may himself proceed to reexpress them by making his own professionally
educated choice and arrangement of words in the other language.

The scientific reader should also adopt this principle and not be content with
knowing approximately what the foreign-language author has said, but require
of himself that he know it completely and idiomatically in the other, generally
his own, language. To do so, the reader or translator must have at least as
good command of his own language as he has of the foreign language.

Most interlingual technical dictionaries, in addition to presenting language
in incorrect units, contain an overabundance of irrelevant and unnecessary
terms, according to expert opinion cited by Holmstrom (1951). These dic-
tionaries are encumbered by "unwanted" terms, which are ordinary dictionary
words, cognates, and compounds with easily identifiable constituents. In a
sample of seven bilingual technical dictionaries, Holmstrom found the average
content of "unwanted" terms to be 90 percent of the total.

For comparison, Holmstrom's criteria were applied to Huebner's German-
English geologic dictionary (1 939) . It was found to contain as much superfluous
material as in any of Holmstrom's samples. Huebner includes countless terms
that are identical in German and English or that differ only slightly in spelling;
compounds with meanings obvious from their elements; and many words and
parts of compounds that are in the German general vocabulary or common
scientific vocabulary. By strict adherence to the criteria, Huebner's dictionary
would have to be reduced from 25,000 entries to about 3000.

210 FREDERICK BETZ, JR.

On reappraisal of Holmstrom's criteria, it is clear that they presuppose a
higher level of competence in languages than can be expected of scientific
readers. It is not realistic to presume that these readers can analyze word
structure in foreign languages. The "unwanted" words in interlingual diction-
aries may, therefore, be essential to the majority of scientific readers. Trans-
lators, as the other main group of users, should be able to get along without
common words in a technical dictionary, but they are in need of explicit
assistance in deciphering technical terms and their derivatives. The profes-
sional translator typically lacks proficiency in technical vocabularies in any
language, as published translations amply demonstrate.

When Huebner's dictionary is reexamined in terms of its usefulness to readers
with an unequal knowledge of German and English, it emerges as an encyclo-
pedic work of substantial quality and value. Besides its primary function, it
gives information on regional usage of both English and German geologic
terminologv, often with more than a mere word equivalent, and it indicates
incorrect usage of German terms in English. Regrettably, the English-German
part of this dictionary has never been published.

Bilingual dictionaries commonly have two parts, each with one of the
languages as the language of reference. In effect they are pairs of dictionaries
in which the contents of both parts are cross referenced. Ideally all terms used
in one part should appear in the other, but the purpose of the dictionary may
dictate differences. For example, Fischer and Elliott (1950) explain that in
their German and English glossary of geographic terms, the English-German
part "is not intended primarily for the German student who wants to read
English geographical literature, but for the English-speaking geographer who
wishes to establish the German equivalent for an English term." Thus, because
ik many English terms are not easily rendered in German," they "are to be
found only in the English-German part." Also, "Terms practically the same
in both languages are, in general, included only in the English-German part;
English readers usually recognize such words but may hesitate to use them in a
foreign language. This is especially true for English terms used in German . . .
and for terms from another language . . . used in both languages." For the
English-speaking user's benefit, the German-English part is the larger, since
his main problem is to understand German literature, not translate English
into German.

In multilingual dictionaries, additional dictionaries can be handled in the
same fashion as the two languages of bilingual dictionaries. The complete
repetition of all terms as many times as there are languages in the dictionary
produces the easiest work for the largest audience to use. In a variant on this
arrangement, a trilingual geologic dictionary prepared by the military geologic
organization of the German army during World War II 6 contains a German-

5 See Wehrgeologenstab Wannsee in the References.

GEOLOGIC COMMUNICATION 211

French-English part followed by a French-German and an English-German
part. Starting with a French or English term in the second and third parts,
respectively, the German equivalent can be used to find the equivalent in
either of these other languages in the first part. Thus, although intended for
primary users of German and arranged for their convenience, this dictionary
can be used effectively by speakers of English or French.

In all dictionaries discussed so far, the terms are listed in alphabetical order.
This simple arrangement is replaced in some dictionaries by listings based on
some type of association of meanings.

Both editions of the Royal Netherlands Geological and Mining Society's
multilingual dictionary (Rutten, 1929; Schieferdecker, 1959) illustrate a non-
alphabetical arrangement of terms. In this general dictionary the field of
geology is divided into major branches or subjects, which are further subdivided
topically. 6 According to Schieferdecker, who edited the second edition, the
arrangement of terms is genetic. The characterization is appropriate for
various sequences of terms within topical divisions, but it is meaningless for
the dictionary as a whole because many of the terms have no genetic connota-
tion and thus their order can have no genetic significance. Schieferdecker
commends the so-called genetic arrangement as a convenience for the reader
by greatly facilitating the finding of terms. On the contrary, the alphabetical
index is the only convenient key to the terms.

In another example of a nonalphabetical arrangement of terms, Challinor
(1961) lists his terms in a classified index under 48 group headings. According
to Challinor, "within each group an attempt is made to put them into some
logical order." Some sequences of terms are familiar and, therefore, appear
to be logical, but others are arbitrary and wholly personal. In this case the
user can ignore the index and look for terms directly in the alphabetical listing
of the dictionary proper.

Despite the weaknesses of nonalphabetical order, it offers the advantage of
grouping terminology by subjects and topics. The user with time to browse
can obtain a complete picture of terminology of a topic without the distraction
of interspersed unrelated terms. The plan converts a general dictionary into
a series of specialized dictionaries.

The associative order of terms in a dictionary is an indirect approach to the
problem of placing terms in context. The additive effect of sequences of words
with related connotations can create an impression of the places in which they
are used properly. A direct approach is taken by scientific reading exercise

6 The first edition (1929) contains eight primary divisions and 44 secondary; the
second edition (1959) has 74 primary and 107 secondary divisions. The great increase
in primary and secondary divisions in the second edition may be due to various causes
indicated in the previous discussion of English dictionaries. The effect of the increase
is to provide more structure for the allocation of terms in the dictionary.

212 FREDERICK BETZ, JR.

books designed for use in foreign-language instruction. Selections from foreign-
language literature may be translated in their entirety into the user's tongue
and printed beside the original text, or only technical terms and other selected
expressions or phrases are translated and given in footnotes, sidenotes, or
appended vocabularies. These books are not systematic presentations of
terminology*, but are often used in place of dictionaries for hasty acquisition of
a specialized foreign-language vocabulary*.

A direct systematic approach to the problem of coordinating terminology
in two or more languages demands an effort beyond translation and compila-
tion. In the late nineteenth century, de Margerie and Heim (1888) became
concerned by the increasing misunderstanding of terminology of crustal dis-
locations. They undertook to clarify- the concepts of the topic by publishing
parallel French and German versions of their text. Where technical terms
appeared, the columns of text were interrupted and the special French and
German terms were centered on the page. In addition, any English equiva-
lents of the French and German terms were given. The authors also appended
references to sources of terms and examples of usage.

Baulig (1956), inspired by this older work, published a study on the termi-
nology of geomorphology in French, English, and German, with incidental
attention to other languages, especially Arabic. The text is in French. Baulig
divided the subject into many topics, which he discussed systematically in more
than 500 numbered paragraphs. Immediately following the appearance of a
French special term in the text, equivalents in French, English, and German,
and in places dialectal terms and equivalents in other languages, were inserted.
The same term may appear in several places in the volume to account for usage
in different contexts. A combined index refers to the numbered paragraphs,
and the different usages are also signalled by cross references in the text.
Baulig summarized the purpose of his study as follows: to reveal to the French
user the resources, often unsuspected, of his own language; to help him hi
reading and possibly translating foreign-language literature; and to contribute
to a revision of the international vocabulary of geomorphology.

Both studies attest to the ingenuity and scholarship of their authors. They
have produced a combination of monolingual and interlingual dictionaries and
explanatory texts that can be used for a variety of purposes. Baulig may have
dealt with the largest practical grouping of subject matter in his study. The
feasibility of applying this plan to a general dictionary is slight. These studies
support an argument for detailed specialized wordbooks, which can portray
technical language in meaningful units.

Interlingual dictionaries call attention to the seriousness of the problem of
worldwide scientific communication. In the absence of an international
language, the primary reporting in one language is not directly usable by those
lacking a reading knowledge of that language. A recent examination of the

GEOLOGIC COMMUNICATION 213

effects of language barriers in earth sciences was made by Emery and Martin
(1961), who found that the literature contains references mostly "to sources
from the same country as the author, evidently because of familiarity with
language and ease of access." They observed that the second most frequent
source for non-English writers is American and British literature, "probably
because English is the most familiar language of the scientific world."

Language barriers are being lowered, though not eliminated, by the simul-
taneous publication of abstracts in one or more of the other widely used lan-
guages along with the original complete text. Journals and volumes of abstracts
issued periodically in various important languages are invaluable guides to
worldwide literature. The dependence on abstracts of foreign-language litera-
ture is paralleled by the growing reliance on abstracts of literature in the user's
own language. The abstract is becoming an ultimate unit of reading, instead
of serving just as a guide to publications.

An alternative to abstracting foreign-language literature is complete or sub-
stantially complete translation. Whereas abstracts are often prepared by
subject-matter specialists, translations are usually the work of professional
translators. The value of completeness of a translation can be outweighed by
the deficiencies in rendition of technical information. Translation is, there-
fore, a way of overcoming language barriers that should be adopted only if
it can be justified. The primary justification is not the existence of a barrier
but the importance of the literature and the need to have its contents
known.

Today, in the United States, bulk translation of Russian literature confirms
the inability of most American scientists to read Russian. If the translations
prepared so far have established the importance of Russian scientific work,
then the obvious solution for American scientists is to learn to read Russian
and not to depend further on translations. In defense of translation on
a grand scale, the limited opportunities for studying Russian in American
schools could be cited. On the other hand, the existence of widespread op-
portunities for studying foreign languages does not guarantee lasting pro-
ficiency.

Emery and Martin (1961) show that American earth scientists do not report
extensive use of French and German sources, although the literature in these
languages is voluminous. A report on proficiency in foreign languages, pub-
lished by the U.S. National Science Foundation (1961), explains the reason for
this surprising observation. The report is based on an inquiry among 100,000
American scientists, who provided the following information: (a) "84 percent
stated they could read scientific or technical material in a foreign language";
(b) of this group only "30 percent rated their ability as good"; (c) "Only 12
percent of those reading German rated their ability as c good'; in French it
was 9 percent . . ."

214 FREDERICK BETZ, JR;

Although many Europeans are competent in the use of several languages,
the growing number of languages in which scientific literature is written poses
a problem for them, too. In 192?, Salomon in Germany was discouraged by
the growing problem in worldwide communication of geologic knowledge.
He predicted, perhaps not altogether facetiously, that the future geologist
would have to spend his first 80 years in learning languages and then only
could he begin with geology-. Salomon proposed the adoption of Latin as the
language for international geologic communication, but was realistic enough
to anticipate strong opposition. The outlook for the success of this proposal
has long since faded, but the need for an international language is greater
than ever.

The increasing use of English in scientific communication may lead to recog-
nition of English as the international language some day, but there is no
prospect that it will displace all other languages in the immediate future.
Until that day, geologists and other scientists who depend on language to
express much of their knowledge, must keep in mind the grave conclusion
stated by Emery and Martin (1961) that:

. . . science is proceeding in each country largely independently of progress
in other countries. This is ... extremely wasteful of manpower, facilities,
and time, although it does provide independent confirmation of ideas and
techniques. Its existence means that probably a majority of scientists tacitly
assume that it is easier to make independent discoveries than to learn of
other prior work when the latter is reported in an unfamiliar language.

Concluding Remarks

The problems of terminology and its unrestrained growth, of dictionaries
and their contents and arrangement, of modes of communication other than
language, and of interlingual exchange of scientific information are all over-
shadowed by the more perplexing problems caused by the rampant growth
of literature.

Emery and Martin (1961) remarked that:

... a scientist of a given field is faced not only by a flood of publications in
his own language but by similar floods in other languages. Since there is a
limit to his time and ability to absorb this material, the scientist may choose
to specialize in a field so small that he can read most of the literature or he
may choose to cover a larger field and ignore most of the literature, particu-
larly that which is difficult to read or obtain.

The problems only appear to have become urgent in our time.
70 years ago, when the flood of literature was a trickle compared with today's

GEOLOGIC COMMUNICATION 215

outpouring, Archibald Geikie (1897) said much the same 7 :

I am only too painfully aware how increasingly difficult it is to 8 keep pace
with the ever-rising tide of modern geological literature. The science itself
has so widened, and the avenues to publication have so prodigiously multi-
plied, that one is almost driven in despair to become a specialist, and confine
one's reading to that portion of the literature which deals with one's own
more particular branch of the science. But this narrowing of the range of
our interests and acquirement has a markedly prejudicial effect on the
character of our work.

Geikie continues with words of admonition that should be repeated today:

The only 9 consolation we can find is 10 the conviction, borne in upon us
by ample and painful experience, that 11 a very 12 large mass of the geologi-
cal 13 writing of the present time is utterly worthless 14 for any of the higher
purposes of the science, and that it may quite safely and profitably, both as
regards time and temper, be left unread. If geologists, and especially young
geologists, could only be brought to realise that the addition of another
paper to the swollen flood of our scientific literature involves a serious
responsibility; that no man should publish what is not of real consequence,
and that his statements when published should be so clear and condensed
as he can make them, what a blessed change would come over the faces of
their readers, and how greatly would they conduce to the real advance of
the science which they wish to serve. 15

7 The text is quoted from the first edition (1897, pp. 287-288) of Geikie's "Founders
of Geology." Several changes were made in the second edition (1907, pp. 471-472),
which are indicated in the footnotes that follow.

8 Inserted between "to" and "keep" in the second edition: "find time for a careful
study of the work of our predecessors, and also to."

9 "The only" is replaced in the second edition with "There is but slender."

10 "we can find in" is replaced in the second edition with "to be derived from."

11 Inserted between "that" and "a" in the second edition: "in the case of geological
literature."

12 "very" is omitted in second edition.

18 "geological" is omitted in second edition.

14 "utterly worthless" is replaced in the second edition with "of little or no value."

15 "." is replaced in second edition with "!."

216 FREDERICK BETZ, JR.

REFERENCES CITED

American Geological Institute, 1957, Glossary- of geology and related sciences: Wash-
ington, D. C., Am. Geol. Inst., 325 pp.; 2nd ed. (with supp.), 1960, 420 pp. Abr. ed.,
1962, Garden City, X. V., Doubleday, 545 pp.

BAULIG, HENRI, 1956, Vocabulaire franco-anglo-allemand de geomorphologie: Paris
(Soc. d'Edition: Les BeUes Leltres) Univ. Strasbourg, Fac. Lettres, Publ., fasc. 130,
230 pp.

BROWN, R. \V., 1956, Composition of scientific words: Washington, D. C., U.S. Natl.
Mus., The Author, 882 pp.

BURT, F. A., 1949, Origins of geologic terms: Sci. Monthly, vol. 69, pp. 20-22.

CHALLINOR, JOHN, 1961, A dictionary of geology: Cardiff, Univ. of Wales Press, 235 pp.

CLOOS, HANS, 1933, Wie sag ich's meinen Fachgenossen?: Geol. Rundschau, Bd. 24,
pp. 225-228.

DARLINGTON, C. D., 1953, The facts of life: London, G. Allen & Unwin, 476 pp.

EMERY, K. O., and MARTIN, B. D., 1961, Language barriers in earth science: GeoTimes
(Am. Geol. Inst.), vol. 6, no. 2, pp. 19-21.

EVANS, BERGEN, 1962, Comfortable words: New York, Random House, 379 pp.

FAY, A. H., 1920, A glossary of the mining and mineral industry: U.S. Bur. Mines,
B. 95, 754 pp.

FISCHER, ERIC, and ELLIOTT, F. E., 1950, A German and English glossary of geographi-
cal terms: Am. Geog. Soc., Lib. Ser. no. 5, 111 pp.

GEIKIE, ARCHIBALD, 1897, The founders of geology: London, Macmiilan; 1st ed.,
297 pp.; 2nd ed. (1905), 486 pp.

H., E. W., 1962, Book review A glossary of geographical terms: Geochem. News,
no. 3, p. 8.

HIMUS, G. W., 1954, A dictionary of geology: Harmondsworth, Penguin Books, 153 pp.

HOLMSTROM, J. E., 1951, Report on interlingual scientific and technical dictionaries:
Paris, UNESCO, Publ. 884, 35 pp.

HUEBNER, WALTHER, 1939, Geology and allied sciences A thesaurus and a coordina-
tion of English and German specific and general terms: New York, Veritas Press,
Part 1, German-English, 424 pp.

HUMBLE, WILLIAM, 1840, Dictionary of geology and mineralogy: London, Henry Wash-
bourne; 1st ed., 279 pp.; 2nd ed. (1843), 294 pp.

KRUMBEIN, W. C., 1962, The computer in geology: Science, vol. 136, pp. 1087-1092.

MARGERIE, EMMANUEL DE, and HEIM, ALBERT, 1888, Les Dislocations de Pecorce ter-
restre Die Dislocationen der Erdrinde: Zurich, J. Wurster, 154 pp.

OLDHAM, THOMAS, 1879, Geological glossary, for the use of students: London, Edward
Stanford, 62 pp.

PAGE, DAVID, 1865, Handbook of geological terms, geology and physical geography:
Edinburgh, Blackwood, 506 pp.

RICE, C. M., 1940, Dictionary of geological terms (exclusive of stratigraphic formations
and paleontologic genera and species): Ann Arbor, Mich., J. W. Edwards, 461 pp.

RICKARD, T. A., 1931, Technical writing: New York, John Wiley, 337 pp.

ROBERTS, GEORGE, 1839, An etymological and explanatory dictionary of the terms and
language of geology; designed for the early student, and for those who have not made

GEOLOGIC COMMUNICATION 217

great progress in that science: London, Longman, Orme, Brown, Green, & Long-
mans, 183 pp.

RUTTEN, L., ed., 1929, Geologische nomenclator: The Hague, G. Naeff, 338 pp.
SALOMON, WILHELM, 1927, Die geologische Literatur und die Sprachen: Geologic,

Nr. 41, pp. 1009-1013.
SCHIEFERDECKER, A. A, G., ed., 1959, Geological nomenclature: Gorinchem, J. Noor-

duijn, 523 pp.
STAMP, L. D., 1961, A glossary of geographical terms: London, Longmans; New York,

John Wiley (1962), 539 pp.
STOKES, W. L., and VARNES, D. J., 1955, Glossary of selected geologic terms with special

reference to their use in engineering: Colorado Sci. Soc., Pr., vol. 16, 165 pp.
STRUNK, WILLIAM, JR., and WHITE, E. B., 1959, The elements of style: New York,

Macmillan, 71 pp.
TOURTELOT, H. A., 1960, Origin and use of the word "shale": Am. J. Sci., vol. 258-A,

pp. 335-343.
UNESCO, 1961, Bibliography of interlingual scientific and technical glossaries: Paris,

UNESCO, 4th ed. 236 pp.
U.S. National Science Foundation, 1961, Foreign-language proficiency of scientists

reporting to the National Register of Scientific and Technical Personnel, 1960:

Washington, D. C., Nad. Sci. Found., Sci. Manpower B., no. 16, 4 pp.
WASHBURNE, C. W., 1943, Some wrong words: J. Geol., vol. 51, pp. 495-497.
WATZNAUER, ADOLF, 1956, Rritische Bemerkungen zur wissenschafdichen Begriffs-

bildung: Zs. angew. Geol., Bd. 2, pp. 64-65.
WEBER, HANS, 1957, Die Sprache in der deutschen geologischen Fachliteratur der

neueren Zeit: Neues Jb. Geol. u. Palaont., Mh., no. 6, pp. 274-278.
Wehrgeologenstab Wannsee, Aussenstelle Heidelberg, 1943, Wehrgeologisches Worter-

buch Deutsch-FranzSsisch-Englisch: (German Army ),Oberkomm. d.Heeres, AHA/In

Fest (Geol.), 449 pp.
WRIGHT, C. W., 1959, Order and disorder in nature: Geol. Assoc. (London), Pr.,

vol. 69, pp. 77-82.
WUSTER, EUGEN, 1955, 1959, Bibliography of monolingual scientific and technical

glossaries: Paris, UNESCO, 2 vols.

JAMES GILLULY

L\ S. Geological Survey

The Scientific Philosophy of
G. K. Gilbert

The philosophy of G. K. Gilbert has earned a place in this symposium
because of his profound influence upon the thinking of geologists everywhere.
Other American geologists, notably Hall, Dana, Rogers, Darton, Daly, and
perhaps a few others, have contributed nearly or quite as importantly to the
body of our science. But Gilbert gave geology a larger dimension and greater
vigor through his thoughtful discourses on scientific method, the creation and
testing of hypotheses, and the disciplined use of the imagination. Much more
influential, however, than all of his essays on method, was the example set by
his scientific memoirs. The conscientious, imaginative observation and the
judicious weighing of alternatives that these memoirs exemplify have become
the ideals by which most of his successors evaluate geologic research to the
present day. Because these qualities have become the standards of American
geology, many geologists who have never read a line of Gilbert's writings are
modeling their work, to the best of their abilities, upon his.

Gilbert's stature and breadth of vision were recognized in his own day. He
was the president of seven scientific societies, the recipient of four prizes and
medals for scientific accomplishment, and the corresponding or honorary
member of many foreign societies. He was the only man ever to serve twice as
president of the Geological Society of America first in 1892 and again seven-
teen years later in 1909. No one can deny his meriting this unique honor;
he was one of the first truly distinguished American geologists.

Gilbert's great powers would have made their impact felt under any circum-
stances. It is not to denigrate them, then, to point out that his influence was
maximized by the scientific climate of his time. He came on the scene at a
turning point of American scientific history, when our science was just emerg-
ing from the long eclipse that began with the death of Franklin precisely the
right time for him to exercise the greatest influence upon the course of the new
day.

218

THE SCIENTIFIC PHILOSOPHY OF G. K. GILBERT 219

For the first two-thirds of the nineteenth century, American science lagged
far behind our technology and invention. Between the time of Franklin and
that of Willard Gibbs, we produced no scientist of the very first rank unless it
be Matthew Fontaine Maury. As Tyndall deplored in 1872, many of our better
scholars were diverted by lack of support for scientific work either into admin-
istration, like Joseph Henry and William Rogers, or into the teaching of other
subjects. Gibbs, for a time, had to teach Latin rather than either mathematics
or physics, even though Yale paid him no salary. The Drapers, father and son,
were perhaps the best of our astronomers, but the father made his living by
writing history and the son by teaching physiology, in order to get the money
to build their telescopes (Cohen, 1959). 1

After the Civil War this situation began slowly to change for the better.
The great step forward for geology began with the mining boom in the West
and the pressures for exploratory surveys as the railroads expanded across the
sparsely settled frontier. With negligible exceptions, this was the first time in
our history when geologists were able to devote their full time and energies to
geologic research. Certainly it was the first time in our history that a consider-
able group of such men were assembled. To borrow a term from our friends
in physics, it is clear, I think, that there is such a thing as a "critical mass" of
scientists who, by mutual stimulation, are able to do much more as a group
than the same men would have been able to do if each were working alone.
Perhaps this generalization would not apply to the very great geniuses, but I
believe it is easy to show that it applies to most scientists of more normal compe-
tence. Every teacher of graduate students recognizes the application of this
rule to a graduate seminar. The blossoming of American geology in the late
years of the nineteenth century took place not only because the descriptive
geology of half a continent needed to be done and men could spend full time
in doing it, nor because the group assembled was one of great geniuses, even
though I think the average competence to have been very high. It took place
because the great challenge of scientific exploration attracted an unusually able
group of men who could devote full time to research in an atmosphere of mutual
stimulation.

No wonder this generation produced so many men distinguished in geology:
Gilbert, Powell, Button, Emmons, Walcott, Hayden, Cross, Becker, Hague,
Ward, Iddings, and McGee. This list could be considerably extended by adding
the teachers in the new graduate universities who were also a part of the same

1 A hundred years ago many able scientists were diverted from teaching and research
into administration because of lack of funds for science; today the plethora of project
funds is luring far more scientists away from teaching and research into management
and fund solicitation than ever were driven away in the simpler days of the last century.
Plus $a change. . . .

220 JAMES GILLULY

program: Chamberlin of Chicago, I. C. Russell of Michigan, G. H. Williams
of Johns Hopkins, and H. S. Williams of Cornell. That most of these men were
exceptionally able is generally agreed, but it seems to me that the impact they
made upon the science was even larger than their abilities would normally
have justified. The combination of great challenge, mutual stimulus, and the
financial support that enabled them to study many facets of geology, not only
areally but also in depth, made possible the flowering of the science in America.

Of these men, Powell was the man of broadest vision, perhaps the most far-
seeing American statesman of his generation, but there is no question that
Gilbert was the man of preeminent influence on geology. His influence was
not revolutionary, like that of Gibbs in physical chemistry, but was more akin
to that of Richards, our first Nobel prize winner, in setting a high standard
of breadth and critical insight into the problem he studied a standard so
high as to form a permanent criterion by which his successors test scientific
excellence.

Gilbert's philosophy was, of course, a synthesis of ideas and methods built
up by the scientific fraternity over the centuries. The task of science consisted,
for him as for others since Bacon, in the observation of natural phenomena
and the discovery of orderly relations between them.

In several of his lectures on methodology, he specifically disclaimed any
novelty of approach, and, indeed, his philosophy did not contain anything
new. That one must have several hypotheses in mind in order to sharpen ob-
servation, that sound judgment depends upon testing more than one hypothesis,
that the scientist should be reserved as to his own conclusions because every
man tends to be partial to his own brain child all these ideas stressed by Gil-
bert and beautifully illustrated in his papers had been familiar since the days
of John Locke and had been eloquently stated by David Hume a hundred
years before Gilbert wrote.

Because the attitude underlying these ideas is so essential to research, Gilbert
properly laid stress upon them. He also was fully aware of the pitfalls of
semantics and of the danger of mistaking words for ideas, a danger from which
the vast rubbish pile of obsolete geologic terms still fails to discourage our too
numerous neologists. In the 1880's he was alert to the problem of relating
language to meaning. Few others in that day considered the constraints placed
on our thinking by language, although Mach in physics and Pearson in statistics
were working along similar lines. These constraints are a principal concern of
present-day philosophy.

The relation Hume called "cause and effect," Gilbert more noncommittally
called "antecedent and consequent.*' He continually stressed one fact: every
antecedent has not one but many consequents. As he put it (Gilbert, 1886,
p. 286), "Antecedent and consequent relations are therefore not merely linear,
but constitute a plexus; and this plexus pervades nature."

THE SCIENTIFIC PHILOSOPHY OF G. K. GILBERT 221

His work gives innumerable examples of his realization that the very com-
plexities of the antecedents and consequents of a particular phenomenon
furnish powerful ways of evaluating hypotheses relating to it. The more conse-
quences one can recognize as implied by some hypothesis, the more abundant
the opportunities of testing them against the facts; the more diverse the conse-
quences, the more sensitive the test. To quote:

[An investigator] is not restricted to the employment of one hypothesis
at a time. There is indeed an advantage in entertaining several at once,
for then it is possible to discover their mutual antagonisms and inconsisten-
cies, and to devise crucial tests tests which will necessarily debar some of
the hypotheses from further consideration. The process of testing is then a
process of elimination, at least until all but one of the hypotheses have been
disproved.

In the testing of hypotheses lies the prime difference between the investi-
gator and the theorist. The one seeks diligently for the facts which may over-
throw his tentative theory, the other closes his eyes to these and searches only
for those which will sustain it. (Gilbert, 1886, p. 286)

This method of simultaneously confronting several hypotheses with the
pertinent phenomena, both antecedent and deduced, goes back at least two
centuries and perhaps, even further, back to Bacon. But it was much in the
minds of both Gilbert and Chamberlin, whose much-quoted paper on the
method of multiple working hypotheses was published ten years later.

Gilbert's method embodies four essential elements: (I) observing the phe-
nomenon to be studied and systematically arranging the observational data,

(2) inventing hypotheses regarding the antecedents of the phenomenon,

(3) deducing expectable consequences of the hypothesis, and (4) testing these
consequences against new observations.

Of the mental qualities involved, Gilbert ranked highest the ability to create
hypotheses. Keen and pertinent observation in the gathering of data itself
demands several hypotheses in order that the mind may be focused on sig-
nificant features. Indeed, hypothesis is virtually implicit in judging what
features of a phenomenon are really significant. An observer cannot observe
everything. If enough possibilities do not occur to him as he is observing, he
may fail entirely to recognize relations of the utmost significance. The bias
arising from the inadequate observation will then vitiate all the logical processes
to which the data may later be subjected.

Gilbert thus stressed highly the value of the creative imagination. He said,

The great investigator is primarily and preeminently the man who is rich
in hypotheses. In the plenitude of his wealth he can spare the weaklings
without regret; and having many from which to select, his mind maintains
a judicial attitude. The man who can produce but one, cherishes and cham-
pions that one as his own, and is blind to its faults. (1886, p. 287)

222 JAMES GILLULY

Gilbert tried to look into the nature of the scientific guess considered as a
mental process. He once asked (1886, p. 286) whether it is possible by training
to improve the guessing faculty, and 9 if so, how. His analysis of his own think-
ing led him to the conjecture that most, if not all, hypotheses arise from actual
or fancied analogies. A newly recognized phenomenon reminds a person,
consciously or unconsciously, of another more familiar one whose antecedents
and consequents are thought to be understood. The mind swiftly constructs
hypotheses in which the antecedents and consequences of the new phenomenon
are similarly related to those of the analogue. This important analysis of
mental processes had of course been suggested long before by Locke and Hume
and had been much emphasized by John Stuart Mill.

Gilbert did not go on in this paper to expand explicitly upon how one is to
develop the capacity to recognize abundant analogies among natural phe-
nomena and thereby become fertile in hypotheses. But one may read between
the lines of many of his papers and learn that diversified experience, both first-
hand and vicarious, was an important catalyst in his own case. One may thus
value the widest possible reading as an aid to the creation of hypotheses in the
sciences of natural history. Reading furnishes experience, even though second-
hand, of a wider range of natural phenomena with which to conjure analogies
than could ever be acquired autonomously by a single person. It is impossible
to reduce most natural relationships to the shorthand of mathematical equa-
tions. Perhaps this accounts for the fact that nearly all studies of scientific
productivity show that geologists, botanists, and zoologists in general attain
recognition later in life than do chemists, chemists later than physicists, and
physicists later than mathematicians. The more complex the phenomena dealt
with, the longer the apprenticeship that must be served before mastery. Cer-
tainly Gilbert's own growth as an investigator continued until his final illness.

Gilbert was no armchair theorist, but an intensely practical working geolo-
gist. He recognized that in our science we must always be satisfied with "as
if." Although by implication he probably had the same attitude toward the
more exact sciences, he expressed this outlook only with reference to the sciences
of natural history. The impossibility of conducting experiments on a geologic
scale demands that we must always be satisfied with hypotheses whose validity
rests on the fact that they have not yet been proved wrong. Excluding directly
observable processes, and even there within embarrassingly narrow limits, we
can never assert that a given phenomenon could have arisen only in a single
way, but merely that it might so have arisen.

If not influenced by personal motives, the mind naturally prefers the simplest
among innumerable possible alternatives. This does not mean, however, that
the simplest theory is necessarily correct. Occam's razor is a mental tool,
not a touchstone. The simplest is only the momentarily preferred hypothesis
and is lightly held pending exhaustive tests. The more numerous the tests

THE SCIENTIFIC PHILOSOPHY OF G. K. GILBERT 223

that an hypothesis survives and the more abundant the uncontradicted deduc-
tions from it, the greater its credibility. Few hypotheses in geology' remain
simple, after analysis. But in the nature of geologic evidence, a geologic con-
cept, even if it survives enough tests to have the rank of theory*, can never be
proved. On the other hand, a single definite negation is enough to disqualify it.
How perilous the way of the geologic innovator!

If, then, the essential elements in Gilbert's philosophy were not new, what
were his personal contributions that left so strong an impress on American
geology? It seems to me that it was less his philosophical essays themselves than
the philosophy he revealed by his superb geologic writings. His influence
derived from his fertile imagination, from the logic and clarity with which he
presented both his conclusions and the chain of reasoning by which he was
led to them, and, not least, in the judicial and essentially undogmatic attitude
he retained toward even his own most strongly fortified conclusions.

To detail the reasoning behind a conclusion seemed to him quite as important
as the conclusion itself. In fact, one of his most influential papers presented no
conclusion at all it merely reviewed the lines along which the problem of the
origin of Coon Butte (now Meteor Crater, Arizona) had been attacked (Gilbert,
1896). He repeatedly stated that review of methods is more valuable than the
actual conclusion reached. An hypothesis that one researcher rejects as in-
adequate may suggest a useful application to another, either in the same prob-
lem or elsewhere.

It is thus through his many beautifully reasoned papers that Gilbert made so
large an impact upon the science. He presented the facts unfavorable to a
tentative thesis quite as prominently as he presented those favorable to it. He
followed the logical implications of a phenomenon wherever they might lead,
holding himself rigidly impartial.

Of innumerable examples in his work, none is more illustrative of his per-
sistence in tracing the consequences of an hypothesis than the last paper pub-
lished during his lifetime (1917). Starting with an analysis of the effect of the
hydraulic mining debris from the Sierra Nevada on the regimen of the streams
in the Sacramento Valley, he found that all of the sediment was trapped at
least 50 miles northeast of the Golden Gate. Few geologists assigned to analyze
such a problem would follow the subject beyond the point where the fate of all
the debris had been determined. Not so with Gilbert. Other consequences
were implied; he felt compelled to follow them out.

The several cubic miles of gravel carried to the upper bay were deposited
below the level of high tide. They therefore filled several cubic miles of space
that formerly had been filled at high tide by water. The tidal prism, the volume
of water flowing out on the ebb, was diminished by a very notable fraction.
The regimen of the offshore bar at the Golden Gate was therefore affected by
the Sierran gravel, even though no gravel had come within 50 miles of the

224 JAMES GILLULY

Golden Gate! The bar had migrated inshore by nearly a mile in fifty years
because of weakened tidal currents. Truly a plexus of antecedents and conse-
quents pervades nature !

Through all his work, as his biographer, William Morris Davis (1926, p. 1)
said, "It was his habit in presenting a conclusion to expose it as a ball might
be placed on the outstretched hand not gripped as if to prevent its fall, not
grasped as if to hurl it at an objector, but poised on the open palm, free to roll
off if any breath of disturbing evidence should displace it; yet there it would
rest in satisfied stability. Not he, but the facts that he marshaled, clamored
for the acceptance of the explanation that he had found for them." Here, in
his example, rather than in the novelty of his philosophy, is the reason for the
tremendous impact of G. K. Gilbert upon American and world geology. 2

REFERENCES CITED

COHEN, I. B., 1959, Some reflections on the state of science in America during the
Nineteenth Century: NatL Acad. Sci., Pr., vol. 45, pp. 666-676.

DAVIS, W. M., 1926, Biographical memoir-Grove Karl Gilbert, 1843-1918: NatL
Acad. Sci. Biog. Mem., 5th Mem., vol. 21, 303 pp.

GILBERT, G. K., 1886, The inculcation of scientific method by example, with an illustra-
tion drawn from the Quaternary geology of Utah: Am. J. Sci., 3d ser., vol. 31 (whole
no. 131), pp. 284-299.

, 1896, The origin of hypotheses, illustrated by the discussion of a topographic

problem: Science, n.s., vol. 3, pp. 1-13.

-, 1917, Hydraulic-mining debris in the Sierra Nevada: U.S. Gcol. Survey, Prof.

Paper 105, 154 pp.

2 In preparing this discussion I have talked over Gilbert's work with many friends;
in this way, I gained many of the ideas here included. Paul Averitt, W. H. Bradley,
T. A. Hendricks, C. B. Hunt, W. T. Pecora, and G. D. Robinson have been especially
helpful.

J. M. HARRISON

Geological Survey of Canada

Nature and Significance of
Geological Maps

A geological map may be defined as one on which are shown the distribution
and structural relations of rocks. As such it is the product of research under-
taken in the geologists' principal laboratory, the earth itself. And because it
describes and interprets the earth, the map must, in the last analysis, be the
source of geological theory.

Many text-books and papers have been written on the technique of present-
ing geological data on maps, and the professional lives of many geologists are
devoted to the preparation and interpretation of such maps. Nevertheless, the
opinion is widely held that geological mapping is a routine procedure gkjn to
an instrumental survey. Such a concept is extremely misleading. Unlike topo-
graphic maps, which record data that can be gathered largely by mechanical,
instrumental, or routine methods, the geological map, although in part objec-
tive and a record of actual facts, is also to a very large degree subjective, because
it also presents the geologist's interpretation of these facts and his observations.
A good geological map is the result of research of a high order.

The first geological map is variously ascribed to Packe in 1743, to Guettard
in 1746, to William Smith in 1801, and to several others, depending on the
author's concept of what constitutes a geological map. The first person to
study geology by making a map appears to have been Guettard, who, in
presenting a paper to the Acad6mie Royale Frangaise in 1746, used a geological
map to demonstrate that England and France were part of the same geological
region (see Mather and Mason, 1939, pp. 77-78). According to Sir Archibald
Geikie (1901, p. 22): "These maps, so far as I know, were the first ever con-
structed to express the superficial distribution of minerals and rocks. The
gifted Frenchman who produced them is thus the father of all the national
Geological Surveys . . ." Geikie also noted that, brilliant as were the deduc-
tions of Desmarest in 1771 concerning the volcanic origin of Auvergne, his
geological map of the region must be considered his most memorable contribu-

225

226 J. M. HARRISOX

tion to the progress of geology (p. ~5'. Most geologists agree, however, that
William Smith's map of England and Wales, published in 1815. marked the
real beginning of geological mapping. Tomkeieff (1948, p. 256) stated that
prior to the publication of Hutton's ''Theory of the Earth" in 1 7 88, or to the
appearance of William Smith's geological map, geologv did not exist as an
organized science. Since that day countless geological maps have been pre-
pared and many areas have been surveyed several times, but great land areas
of the world are still unmapped, and the geology of the sea floors is virtually
unknown.

Am map of the surface of the earth can convey some information on the
geology of the area. Air photographs, and the topographic maps derived from
them, may show surface characteristics due to faults, folds, varying lithology,
and ocher features, so that a good first approximation of the geology of the area
can be obtained by an observer who is familiar with the interpretation of such
features. Maps resulting from airborne magnetometer surveys may be useful
as interim substitutes for geological maps in areas where geological observa-
tions are lacking, and, of course, they are widely used to supplement geological
data. Similarly, soil maps and geochemical maps are useful guides to geology.
All such maps disclose much about the geology of an area because, in most
regions, it is the bedrock that controls the geochemical and geophysical charac-
teristics.

To interpret geology from such maps, a geologist need not see a single rock
or rock outcrop. His interpretation is made by applying his geological knowl-
edge and experience to the study of chemical and physical data, topographic
forms, colour variations, etc.; he deduces from these observations the physical
cause of the observed phenomena. Although these maps are important to a
geologist, they are not geological maps in the true sense because they do not
portray the geology of the area geological information is only incidental.

For most people, the geological map, with its scheme of contrasting colours,
apparently unequivocal structural symbols, and sharply drawn contacts be-
tween rock units, creates the impression that it is, like most other types of maps,
a factual and objective record of data derived from observations made on
different classes of rocks clearly distinguishable from each other by well-defined
physical characteristics. For most geological maps this impression is fallacious.
A good geological map is much more than an objective presentation of the
distribution of rock units, their structure and their relations; it is also a sub-
jective presentation of interpretations based on a multitude of observations and,
to a greater or less degree, based on theories and prejudices held at the time
the map was made. This is true of the first maps made by Guettard, Lehmann,
Packe, Smith and other pioneers, and it is true of the maps made today. We
have accumulated much more knowledge since the days of the pioneers, and
we hope we are producing maps more closely approximating the truth. But

NATURE AND SIGNIFICANCE OF GEOLOGICAL NL^PS 227

geological maps are not static or timeless; as the science evolves, so will the
maps that portray our evolving concepts.

Consider the preparation of a modern geological map. \Ve may assume the
geologist has an accurate topographic map and air photographs, and that he
will assemble all available geophysical, geochemical and other useful data
pertaining to the area. From these he makes his first interpretative study and
produces a rough outline of the geology. With it he plans critical traverses
and selects sections for detailed study. Only then will he begin examination of
the rocks and deposits themselves, including classification of the different rocks
according to their physical characteristics and his interpretation of their origin.
He must, in other words, decide on his classification before he can show it on
the map. He must decide, for example, whether the rock is magmatic or mig-
matitic, whether it is volcanic or intrusive, whether two rocks represent differ-
ent formations or different facies of the same formation, whether the rock was
formed from a volcanic, marine, or continental sediment, whether it is meta-
morphic or metasomatic, and so on. The fact that the classification must be
made in the field is probably unique in the scientific disciplines. It would
indeed be a rash geologist who, in an area of even moderate complexity, would
make a geological map based on a collection of hand specimens.

Not only is he concerned with the classification of the rocks, but also with
the structural data relative significance of unconformities, relationship of
cleavage to folds and faults, relative movement on faults, age of structural
deformation, age of igneous intrusions, and all the complexities of rock rela-
tionships. These data, together with those available from geophysics and geo-
chemistry, borehole records, and other sources will be used to extend lithologic
and structural units beneath the overburden. Even in areas where rocks are
relatively well exposed, only a small part of the bedrock crops out, and these
croppings, moreover, are usually not uniformly distributed.

The geologist always collects much more information than he can show on
the map, so he must select the data he considers most significant. For example,
he determines the mineral assemblages in scores of thin sections so that he may
refine his classification of the rocks, or distinguish half a dozen metamorphic
facies. These classifications, in turn, lead to conclusions concerning the geo-
logical history of the area. He may make a statistical analysis of hundreds of
field observations to determine the shape of folds, the position of faults, direc-
tions of paleocurrents in ancient seas and rivers, or the significance of fossil
assemblages. These investigations constitute the author's laboratory research,
and he includes his interpretation of the results on his map along with his inter-
pretations of geochemical analyses, geophysical measurements and isotope
abundances.

Finally, the geologist must select the symbols and the scheme of presentation
for all this information so that the map will best portray his selection of data

228

J. M. HARRISON

A V>

Geologically mapped 1928

Batholithic intrusions;
granitic rocks, locally
numerous inclusions
of Grenville series

Basic intrusions

Crystalline limestone,
quartzite, garnet gneiss;
locally abundant intrusions
of granite

Scale of miles

01234

i | i | i

FIG. 1. Two geological maps of the same area in the Canadian Shield.

NATURE AND SIGNIFICANCE OF GEOLOGICAL MAPS

229

Geologically mapped 1958

Granitic rocb

Basic rocks,
mainly intrusions

Granitized rocks,
migmatites, etc.; includes
some granite

Crystalline limestone,
quartzite, paragneiss;
includes some granitic and
granitized rocks

Scale of miles
1 2 3

FIG. 1 (Continued}

230 J. M. HARRISON

and interpretations. They must permit the user to look into the map, so that he
may visualize the distribution and relationships of the rocks beneath the surface
of the earth Olackin, 1950, p. 55).

When all this material is synthesized, the final product is a lithographed
geological map that looks as positive and incontrovertible as a colour photo-
graph, at least to the layman and perhaps also to some geologists. Actually, it
represents the sum of analyses made by an individual, using information that
never can be complete. The results are conditioned by the "conventional
wisdoms" of the day and to that extent the map represents the "geological
knowledge at the time of its production" (North, 1928, p. 1). Moreover, in
many instances, more than one interpretation can be made of the information
presented on the map; another geologist may make interpretations quite
different from those of the author and may even see more than the author
realized was there.

In most regions, geologists of the same generation would produce similar
maps, but if the origin of certain rock units were controversial, and the geol-
ogists were products of different schools of geological thought, their maps would
be different. For example, the area of "granite" shown on a map of a meta-
morphic terrain could depend largely on whether the author belonged to a
school believing that granites are, in the main, of magmatic origin, or to a
school believing they are the product of ultrametamorphism or granitization.
This may be an extreme example, but most geological maps to a greater or
lesser degree reflect the background and prejudices of the author. Good maps
also bring to light weaknesses in theory-, or in the interpretation of experimental
geology-, or in classification. In other words, they lead to their own modifica-
tion and eventual obsolescence. Certainly later generations of geologists will
produce maps different from those of today.

The field geologist, besides providing basic information, is in a good position
to appraise theories developed from laboratory work. Unhappily, many field
investigators tend to accept the results of laboratory- experiments with immod-
erate faith and to modify their mapping to bring it into accord with experiment
and theory. The field man should not forget that the laboratory investigator,
in his attempts to simulate geological processes, must extrapolate on a grand
scale and thus may make unwarranted assumptions.

An example that comes to mind involves one of the great men of geological
science, N. L. Bowen. About 1915 he began publishing the most outstanding
petrological contributions of his generation, which culminated in his "The
Evolution of the Igneous Rocks,* 9 published in 1928. His theories were based
almost exclusively on brilliant laboratory research, and they dominated the
thinking of nearly all geologists in North America for a generation or more.
Too few geologists tested these concepts in the field; many geologists mapped
according to them, overlooking all contradictory evidence. It was not until

NATURE AND SIGNIFICANCE OF GEOLOGICAL MAPS 231

1947 that enough heat was generated by field geologists to make a symposium
on granites the major event at the Annual Meeting of the Geological Society
of America in Ottawa (Geol. Soc. Am., Mem., 1948). Partly as an outcome
of that meeting, the theory of granitization once more gained recognition among
North American geologists, but only because the field evidence demanded its
acceptance in opposition to prevailing theory.

For an example of how theory can control mapping see Fig. 1 . The two
maps, greatly simplified from the originals, are of the same area in the Canadian
Shield. The older one was prepared when "magma' 5 was at its apex in North
America, the later one by a geologist who had accepted the field evidence for
granitization. Part of the discrepancy is due to the more modern realization
that some hornblende-rich and pyroxene-rich rocks have been derived from
limy sediments. The basic difference, however, is due to the fact that the more
recent author shows on the map his conclusions about the origin of the various
rocks. Maps that more closely resembled the later edition of these two had
been published for other areas by many geologists over a long period. It was
the weight of their evidence that made it necessary to accept, in opposition to
laboratory theory, the principle of large-scale granitization.

Similarly it was many years before field geologists and laboratory specialists
resolved their differences on the temperatures of molten rocks. Laboratory
investigations of silicate melts apparently demanded extremely high tempera-
tures, but field geologists insisted that evidences of such temperatures could
not be confirmed where magmatic material had invaded solid rocks. Eventu-
ally, it was found that addition of small amounts of "mineralizers/ 3 mainly
water, reduced the temperatures of the melts to the point where they were in
accord with the evidence from the earth itself.

To complete North's concept "The geological map is an index of the extent
and accuracy of geological knowledge at the time of its production, and is also
the basis of future research. It is the vehicle by which men communicate to
one another their discoveries relating to the nature and arrangement of the
rocks of the earth's crust, [and it] makes possible the prosecution of further
research concerning the distribution of rocks, their origin and the evidence
of the life of the past which they may contain . . . The geological map may,
therefore, be regarded as the dynamic force in geology." (North, 1928, p. 1)

232 J. M. HARRISON

REFERENCES CITED

BOWEN, X. L., 1928, The evolution of the igneous rocks: Princeton, Princeton Univ.

Press, 332 pp.
GEIKIE, Sir ARCHIBALD, 1901, The founders of geology, vol. 1 of Principles of geology:

Baltimore, Johns Hopkins Press, 297 pp.
Geological Society of America, 1948, Origin of granite (James Gilluly, Chm.): Geol.

Soc. Am., Mem. 28, 139 pp.
GREENLY, EDWARD, and WILLL\MS, HOWELL, 1930, Methods in geological surveying:

London, Thos. Murby, 420 pp.
IRELAND, H. A., 1943, History of the development of geologic maps: Geol. Soc. Am., B.,

vol. 54, pp. 1227-1280.

LINTON, D. L., 1948, The ideal geological map: Adv. Sci., vol. 5, no. 18, pp. 141-148.
MACKIN, J. H., 1950, The down-structure method of viewing geologic maps: J. Geol.,

vol. 58, pp. 55-72.

MATHER, K. F., and MASON, S. L., 1939, Source book in geology: New York, McGraw-
Hill, 702 pp.
NORTH, F. J., 1928, Geological maps, their history and development with special

reference to Wales: Cardiff, Nat. Museum, Wales, 133 pp.
POUBA, ZDENEK, 1959, Geologicke mapovani: Nakladatestvf, Ceskoslovensk6 Akad.

Ved, 523 pp., 12 maps. (A complete account of the technique of geological mapping.)
ROBERTSON, THOMAS, 1956, Presentation of geological information in maps: Adv. Sci.,

vol. 13, no. 50, pp. 31-41.
STEINER, W. VON, 1957, Zur Geschichte der geologischen Karte: Zs. Angew. Geol.,

nos. 8/9, pp. 417-424.
TOMKEIEFF, S. L, 1948, James Hutton and the philosophy of geology: Edinburgh Geol.

Soc., Tr., vol. 14, pt 2, pp. 253-276.

WELLS, J. W., 1959, Earliest geological maps of the United States, 1756-1832: Washing-
ton Acad. Sci., J., vol. 49, pp. 198-204.
WOODWARD, H. B., 1911, History of geology: London, Watts, 154 pp.

ARTHUR F. HAGNER

University of Illinois

Philosophical Aspects of
the Geological Sciences 1

The geologic approach to nature raises some interesting philosophical ques-
tions, and at the same time exemplifies a point of view which should contribute
to the thought and progress of other fields, if it can be clearly delineated. An
attractive area of knowledge, still relatively unexplored, invites analysis and
evaluation.

Intellectual Contributions of Geology

In the eighteenth and nineteenth centuries, geologists generated concepts that
commanded universal attention. A venerable belief in an earth no older than
a few thousand years, formed almost instantaneously by providential action,
had long suppressed philosophic and scientific thought. Prior to the time of
Hutton, moreover, it was generally believed that this brief past had been
marked by extensive and violent changes, unlike any taking place today.
Surface features of the earth were explained in terms of the Deluge. Erosion,
the effects of glaciation in regions where today the climate is warm, and the
presence of fossils on mountains were all attributed to Noah's flood. Geologic
discoveries, and the resulting controversies in science and religion, led to a
gradual abandonment of these beliefs.

Before James Mutton's "Theory of the Earth" appeared in 1788, natural
philosophy was based more on speculation than observation. Hutton's uni-
formitarianism made it possible to explain earth features as a result of long-
continued but ordinary processes, as opposed to extraordinary forces or cata-

1 I would like to thank Dr. D. M. Henderson for carefully reading the manuscript
and for offering suggestions that materially improved it. Dr. Norman Page contributed
time and suggestions during the early stages of writing and preparation. A number of
the authors of this volume have furnished helpful comments and criticism. I also wish
to thank those graduate students in geology at the University of Illinois who took part
in the informal and stimulating discussions on philosophy of science that were held
during the academic year 1960-1961.

233

234 ARTHUR F. HAGXER

clysms. Speculations slowly gave way to careful observations, correlations, and
substantiated interpretations. It was not, however, until the publication of
LyelTs "Principles of Geology"' in 1830-1833 that the revolution in this science
became widespread and effective. By the middle of the nineteenth century
fc a satisfactory geological philosophy in the inorganic world had been attained
and established on a solid foundation of observation, so far as the criteria of
the times would allow. In the organic field the bomb-shell had not yet burst."
(Gordon, 1951j

The idea of progressive change and development with time, as applied to
physical science, has matured \\ith the gro\\th of modern geology. Previously
historians had discussed the idea of progress, but onlv casually. When Hutton
spoke of seeing no signs of a beginning or of an end, he realized that time was
needed to produce many geologic phenomena more time than scientists and
philosophers were prepared to concede. The idea of geologic time enabled
Darwin to construct his theory of evolution on a scientific foundation. "The
principle of Uniformitarianism had to be clearly established, or else the history
of life of the past could have a thousand interpretations based on physical
conditions entirely dissimilar to those of the present/' (Garrels, 1951) It was
LyelFs Principles, then, that set the stage for Darwin and greatly influenced his
ideas. "A long history of life on the earth, undisturbed by repeated world-wide
catastrophies, no longer required successive creations of all living organisms . . .
The possibility of the progressive development of life on the earth could no
longer be dismissed/' (Thomas, 1947; The idea of geologic time, perhaps the
greatest contribution of geology to general thought, has been compared in
importance with the astronomers* realization of the vastness of space and with
the physicists* concept of the relationship between matter and energy. (Stokes,
1960, p. 5)

Geology thus made a major contribution to both science and philosophy by
introducing the idea of history into science. Prior to the nineteenth century,
science had been concerned largely with the present. Not until geologists
introduced the concept that the earth has a history was it possible to develop
a systematized knowledge of the remote past. Geology demonstrated that it is
possible to study the past by scientific methods and that there "is a validity to
history apart from and independent of physics and chemistry- . . . The theory
of evolution was a historical idea, not a law of nature. Its validity was to be
tested by other criteria than those of mechanics." (Schneer, I960, pp. 377-378)

Characteristic Features of Geologic Methodology and Reasoning

In discussing the nature of geology-, Chamberlin (1904) observed that "Not
a little consists of generalizations from incomplete data, of inferences hung on
chains of uncertain logic, of interpretations not beyond question, of hypotheses

PHILOSOPHICAL ASPECTS OF THE GEOLOGICAL SCIENCES 235

not fully verified, and of speculations none too substantial. A part of the mass
is true science, a part philosophy ... a part is speculation, and a part is yet
unorganized material. 33 This remains true today, although the science is
generally more quantitative and its generalizations and interpretations stand
on a somewhat firmer basis of fact. Nevertheless many geologic theories rest
on what Chamberlin (1 904 ) referred to as "the working test"' and Geikie ( 1 905)
called "a balance of probabilities/ 3 Often the geologist's explanation turns on
his judgment in selecting the most likely working hypothesis and revising it as
data accumulate. Consequently, as stressed by Bemmelen (1961 j, the personal
capacity of the scientist is more important here than in sciences which rely
mainly on instruments and do not use an historical approach to their problems.
Geology has had to advance more by observation, description, and classifica-
tion than by experiment and calculation.

The geologist has to work in an intellectual environment to which many
other physical scientists are unaccustomed. In order to understand the nature
of geologic problems and concepts, one must learn to live with uncertainty to a
degree not imposed by problems involving closed systems, isolated variables,
verifiable experiments, and the statistical treatment of large numbers of ob-
servable occurrences. Of the physical scientists, moreover, the geologist is most
often confronted with the problem of working with end products, which result
from the interplay of many complex variables. It is usually difficult to isolate
the variables for realistic experimentation. Commonly there are more variables
than the number of parameters needed to reach a solution, and these operate
over extreme ranges of magnitude.

Geology differs from other sciences mainly in its concern with time. Processes
and reactions that may take place so slowly as to be practically unobservable
in the laboratory may be of great importance when given many millions of
years of operation. Often the geologist must try to determine whether a phe-
nomenon resulted from an intense process that acted through a short time, or
whether the process operated at lower intensity over a longer period. Can he
directly relate short-term laboratory experiments to natural phenomena that
may have required an almost infinitely greater amount of time? Is he justified
in speeding up the rate of an experiment in order to reduce the time factor?
The geologist must rely not only on the soundness of his experiments, but also
on the realization that, given enough time, many phenomena experimentally
unreproducible at present can materialize in nature. A positive mental effort
is always required to view problems in their proper temporal perspective.

Another major difference between the viewpoint of the geologist and that
of other scientists is that the former commonly finds it necessary to "predict
the past." Bubnoff (1959) believes that the principal difference between
geology and other physical sciences "lies in the fact that geology not only must
explain the contemporary situation by means of contemporary phenomena,

236 ARTHUR F. HAGXER

but it must also observe and point out a genesis, a process which takes place
in time and of which the different stages cannot be reached by us through
immediate observation." Umbgrove (1947) has compared the geologist with
the historian who reconstructs the past by using available data, and where
these are incomplete s employ's temporary constructs to bridge missing data.
"Since many earth phenomena cannot be verified experimentally, and com-
monly the geologist cannot 'get at' the problem directly, it is necessary for him
to use various indirect methods of analysis . . . These methods of reasoning
and 'explaining 5 are necessary because the geologist deals with time and scale
factors beyond human experience." (Hagner, 1961)

Physicists and chemists are accustomed to contemplate and manipulate
exceedingly minute objects. While this is true to some extent of geologists,
they must also attempt to conceive of the relative size and behavior of units
of matter much larger than the observer. One may observe the tracks of atoms
in a cloud chamber, but it will probably be some time yet before man can ade-
quately observe a continent in motion. Scale hi geology ranges from the sub-
microscopic to the planetary, from the structure of crystals to the structure of
the earth.

In geology there is a rapidly growing trend toward quantification; neverthe-
less, geology differs from some of the other sciences in the lesser degree to which
quantitative data are available. In this connection, many scientists seem to
believe that qualitative data are wholly subjective, whereas quantitative data
are objective. But, as Birch (1951) has said, "No one ever discovered a quality
apart from a quantity nor a quantity apart from a quality. Why, then, are we
so anxious to adopt the weird hypothesis that the quantitative is objective and
real but the qualitative only subjective?" Geologists are deeply interested in
this matter because they must deal with an "indispensible core of qualitative
observation that forms the foundation of virtually every geological study . . ."
(Krumbein,1960)

As already stated, the physicists and chemists have been concerned largely
with problems of the present and with a time scale closely related to that of
the observer. It has been said that these sciences have been successful in the
main because they have deliberately restricted their temporal scope, thus avoid-
ing the kinds of problems that cannot be approached by experiment. The
geologist, however, cannot escape such problems and, despite the complexity
of earth phenomena and the incomplete record of earth history, he has suc-
ceeded largely by qualitative methods in establishing concepts which have
withstood the test of time and which, in some cases, have been verified experi--
mentally.

Perhaps the most important factor in the progress of geology has been the
development of a "geologic frame of mind." This is acquired almost of necessity
in the day-to-day effort of thinking about immense periods of time, very large

PHILOSOPHICAL ASPECTS OF THE GEOLOGICAL SCIENCES 237

units of matter, and the interplay of complex variables. Since it is impossible
in geology to duplicate nature in the laboratory, natural phenomena are
commonly studied as units. Where possible, factors are isolated for analysis,
but since it is impossible to bring a mountain into the laboratory, experimenta-
tion plays a less important role than it does in physics and chemistry. The
study of nature as a whole calls for the kind of reasoning in which all the vari-
ables are considered insofar as possible, even though this makes it necessary
"to supplement in thought partially available facts." (Mach, 1903) The co-
ordination of directly observable data with unobservable data by mental sup-
plements has yielded models sufficiently useful to predict the location of sub-
surface oil, gas, and mineral occurrences.

Although interpretations of the evolution of mountains, basins, and the crust
of the earth itself must be based on physical laws, there are many things con-
nected with these and other gross features of the earth "that cannot be squeezed
into a formula and can only be described." (Bertalanffy, 1952) As Bemmelen
(1961, p. 458) has said,

. . . matter reacted upon matter in an infinite number of combinations
in such a way that ultimately new possibilities and new factors originated,
so-called 'emergent phenomena.' The latter cannot straight away be ex-
plained by the natural laws of the basic sciences . . . the natural sciences
tend to possess a certain type of hierarchy in which the rules and laws of
the simpler stages are also valid for the higher organized ones, but not
vice versa.

Because the earth does not behave like a stone, geologists have proceeded on
the basis that, unless truly fundamental and pertinent physical laws are vio-
lated, geologic evidence must be accepted at face value. "In many instances,
such as the controversies in the nineteenth century over the age of the earth
and the thickness and rigidity of its crust, history has shown the geologists to
have been more nearly correct than the other physical scientists." (Hagner,
1961)

Geologic concepts and theories differ greatly in the completeness and quan-
tity of data or observations on which they are based, and consequently in their
reliability and general acceptance. The theory of magmatic differentiation by
crystal settling and crystal fractionation, for example, is well substantiated
both by field observations and by experimental evidence. For any theory of
mountain-building, however, there is no general agreement, because the data
on which it rests are fragmentary. Consequently there are few rigorous laws
applicable to large-scale earth phenomena.

Physics and chemistry deal largely with processes and the prediction of
results from the action of these. For example, a basic assumption of thermo-
dynamics is that it is necessary to know only the state of a system and the condi-

238 ARTHUR F. HAGXER

tions imposed on it in order to predict the result no knowledge of the "path"
of the process is required. Geology deals with macroprocesses made up of many
subsidiary dynamic processes. The observable end product is the sum of serial
microprocesses that operated along a particular, often tortuous, path. The
study of history in science is the diagnosis of pathways. Many lines of geologic
evidence suggest that two different paths may lead to approximately the same
kind of end product, e.g., the formation of granite by different processes. Such
a possibilitv runs counter to a belief, perhaps intuitively held in some sciences,
that distinctive end products generally form in onlv one way.

It is interesting to note that in the twentieth century, physicists have come to
suspect that their reasoning and subject matter have been too restricted. The
immensity of time, for example, has received much attention in the theory of
relativity. Today the theoretical physicist must infer the unobservable. He is
increasingly concerned with phenomena that cannot be transferred into actual
sense perceptions and tested by direct experience. Like the geologist, he must
''supplement in thought partially available facts/' Major advances in physics
and chemistry may well come as a result of concentrating on the whole rather
than its smallest parts. This will call for reasoning more akin to that of the
philosopher, historian, geologist, and biologist than is customary today among
most phvsicists and chemists.

The Position of Geology Today

During the twentieth century a number of new concepts of general applica-
tion and great fundamental importance have arisen both in physics and biology.
These new ideas or principles pertain to all of science, not merely to one par-
ticular area of investigation. Although geology has advanced greatly during
this period, it has produced little or nothing of similar significance. This is
somewhat surprising, for many of the new ideas have long been implicit in
some fields of geologic research and in many lines of geologic reasoning.

Recently emphasis has been placed by certain biologists (Bertalanffy, 1952)
on a whole or organismic concept of nature the organic idea of Whitehead
as opposed to an analytical point of view. This idea stresses the importance of
the relations between the individual parts of an organism or system, and shows
that a complete description of the organism must include the laws governing
these relations as well as the laws governing the behavior of the component
parts. Coupled with this idea is that of structure and form as temporary mani-
festations of the interaction of processes proceeding at different rates. It is the
process that is fundamental, and nature so viewed is dynamic rather than
static. At the same time the concept of open systems has been fitted into this
general scheme, the organism being viewed as a locus of interacting processes
maintaining itself in a steady state of minimal entropy by the constant dissipa-

PHILOSOPHICAL ASPECTS OF THE GEOLOGICAL SCIENCES 239

tion of energy. Ideas of hierarchy of order have been formulated, each level
with its own set of laws transcending those of the lower levels.

These concepts deal with more complex organizations of matter than are
usually considered in physics, but they deny neither the basically statistical
character of the laws of nature in the microphysical realm, nor the discontinu-
ous nature of primary events. Physical concepts have been incorporated into
modern biology, with necessary attention to the biological factors involved.
Even Bohr's complementarity principle, the idea that sometimes nature cannot
be described by one concept but rather by pairs of opposed and complementary
concepts, has found an application in biology. Just as in physics both the wave
and particle descriptions of matter apparently are equally necessary and valid,
so in biology the organismic and physicochemical approaches appear to be
required for understanding.

Geologists have dealt with these concepts in a qualitative manner and in
varying degrees of explicitness, but apparently this has not been fully appreci-
ated by scientists in other disciplines, or even by geologists themselves. Geology-,
too, is particularly concerned with the interaction of processes of different rates,
and it has long been realized that a description of only one process operating
in an area is incomplete and may be misleading unless its relations to other
processes are known. In other words, the concept of the whole is reaffirmed.
There has also been a continuing consideration among geologists of the relative
importance of uniformity and catastrophism in nature. This is evident in such
fields as tectonics, paleontology, and stratigraphy; in fact, when dealing with
geologic periods of time, this consideration can rarely be avoided. In perhaps
no other branch of science is the transitory nature of form and structure so
clearly realized as in geology. Most geologic systems are "open" in that both
matter and energy' may move freely in and out of them, although there is a
constant approach to temporary or local equilibrium. This approach to local
equilibrium is a major unifying concept in geology.

Today, "Geology is in a period of unprecedented discovery . . . and tradi-
tional views concerning the physical and chemical processes that have produced
and are now moulding the earth's crust are now being challenged . . . there is
an atmosphere of fascinated suspense today such as has always marked the
high points on the growth curve of a science." (Bucher, 1950) Major advances
in geology can indeed be expected. Research methods and techniques are
much more precise than those of the past; consequently it has become possible
to analyze physical and chemical processes and phenomena more accurately
than when these were amenable only to speculation or reasoning from inade-
quate qualitative information. Major contributions to our knowledge of con-
tinental growth and movement, the nature of the earth's interior, and the
distribution and movement of material in the earth's crust are to be expected.
Today the heredity principle is being applied to such diverse phenomena as

240 ARTHUR F. HAGNER

tectonism, igneous processes, hydrothermal activity, and the inheritance of
properties by successive mineral phases. Pospelov (1961; has discussed the
concept of multistage geologic complexes as having validity on a global as well
as regional scale. The concept of paragenesis has been enlarged and becomes
bi the basis of the theory of geologic formations, metallogenic complexes, etc.
It pervades the entire problem of relationships between igneous activity and
mineralization, between metamorphism and tectonics/' Pospelov has also
stated that "geology- has come face to face with an understanding of the higher
type of geologic form of motion, i.e., an understanding of the active self-
development of the earth."

In spite of the marked progress of geology during the twentieth century, its
relative position among the sciences has declined. This is partly a result of the
spectacular findings in such fields as quantum physics and molecular biology.
But other factors have contributed to this decline and some of these are attrib-
utable to disinterest and negligence on the part of geologists. Geology, perhaps
more than other sciences, has suffered from a fragmentation into numerous
semi-independent disciplines, so that geology is now essentially a group of
sciences. Concurrently with fragmentation, the boundaries between geology*
and other sciences have been disappearing, and new names have been given
to the interdisciplinary fields. The subject matter of some of these vigorous
hybrids geophysics, geochemistry, microcrystallography has been defined
so as to include the more concrete aspects of geology that can be studied by
mathematical., physicochemical, and statistical methods. What is arbitrarily
left under the label "geology-" thus constitutes the less definite, more uncertain
aspects of the science.

Need for Studies in the History and Philosophy of Geology

Geologists have accumulated vast amounts of data and are faced with much
unorganized material requiring analysis and synthesis. Scientists in other
disciplines are concerned with the possibility of achieving not only synthesis
in their fields, but also a unification of all science. Because geology rests in
part on physics, chemistry, and biology, in addition to being a science in its
own right, the geologist is in an excellent position to appreciate attempts to
unify science and to contribute to them. But geologists have published essen-
tially nothing on this challenging subject. Is this field of knowledge also to
be preempted by other scientists and philosophers to the exclusion of geologists?

Geology must elucidate its uniqueness of approach, demonstrate the ways
it has contributed ideas of interest to all sciences, and formulate and name
these ideas as specific principles or concepts. We must show wherein we deal
with nature in general. We need to be more self-analytical in our processes
of reasoning, as scientists in other disciplines have been. Recently a few studies
have appeared on the problems of sampling, scale, time and measurement in

PHILOSOPHICAL ASPECTS OF THE GEOLOGICAL SCIENCES 241

the geosciences, but long overdue are searching analyses of the nature of
geologic concepts, methods, reasoning, evidence, and interpretation. In other
words, we should take inventory of the nature, subject matter, accomplish-
ments, and directions of geology. There is an obvious need for fundamental
thinking about geology.

Chamberlin (1904) has said "an appropriate atmosphere of philosophy . . .
is necessary to the wholesome intellectual life of our science . . ." Study of the
history and philosophy of science would add "perspective to the succession of
facts as well as methods" and expose the "bond which holds scientific thought
together." (Margenau, 1960) By history is meant the tracing of the develop-
ment and interaction of ideas among different disciplines. The very nature of
geology and its methods of study, in part historical and philosophical, should
encourage the geologist to examine the evolution and substance of his ideas.

REFERENCES CITED

BEMMELEN, R. W. VAN, 1961, The scientific character of geology: J. Geol., vol. 69,

pp. 453-461.
BERTALANFFY, L. VON, 1952, Problems of life, an evaluation of modern biological

thought: London, Watts, 216 pp.

BIRCH, L. C., 1951, Concept of nature: Am. Scientist, vol. 39, pp. 294-302.
BUBNOFF, S. VON, 1959, Grundprobleme der Geologic: Berlin, Akademie-Verlag, 234 pp.
BUGHER, W. H., 1950, The crust of the earth: Sci. American, vol. 182, no. 5, pp. 32-41.
CHAMBERLIN, T. C., 1904, The methods of the earth-sciences: Popular Sci. Monthly,

vol. 66, pp. 66-75.

GARRELS, R. M., 1951, A textbook of geology: New York, Harper, 511 pp.
GEKIE, Sir ARCHIBALD, 1905, The founders of geology: London, Macmillan, 486 pp.
GORDON, W. T., 1951, Geology, HI, H. Dingle, ed., A century of science: London,

Hutchinson's Scientific and Technical Publ., pp. 98-113.
HAGNER, A. F., 1961, Geologic education and its influence on approaches to geologic

problems: J. Geol. Educ., vol. 9, pp. 89-97.
KRUMBEIN, W. C., 1960, The "geological population" as a framework for analyzing

numerical data in geology: Liverpool and Manchester Geol. J., vol. 2, pp. 341-368.
MACH, E., 1903, quoted in Schrodinger, Erwin, What is life? and other scientific essays:

New York, Doubleday, 1956, 263 pp.
MARGENAU, H., 1960, Foreword to, The search for order, by C. J. Schneer: New York,

Harper, xii pp.
POSPELOV, G. L., 1961, Geology as a science and its place in natural history: Izvestiya

Acad. Sci., USSR, Geol. Ser., 1960, (trans. Nov., 1961), pp. 1-11.
SCHNEER, C. J., 1960, The search for order: New York, Harper, 398 pp.
STOKES, W. L., I960, An introduction to historical geology: New Jersey, Prentice-Hall,

502pp.
THOMAS, H. H., 1947, The rise of geology and its influence on contemporary thought:

Ann. Sci., vol. 5, pp. 325-341.
UMBGROVE, J. H. F., 1947, The pulse of the earth: The Hague, Nijhoff, 358 pp.

ROBERT F. LEGGET

Research Council of Canada

Geology in the

Service of Man

From the dawn of history, the development of the human race has involved
close contact with the earth. On the surface of the earth man has made his
home, from the time of the earliest mud hut to the present day \vith its towering
sk\ scrapers providing eyries for urban dwellers. From mines excavated within
the upper few thousand feet of the earth's crust, man has obtained his fuels for
heat and power, his precious metals for barter, his serviceable metals for use
and decoration, and much of the material he has needed for building. Upon
the ground he has developed his transportation routes, from the earliest paths
between forest settlements to the great highways, railways, and airports of
today, boring tunnels through the hills when necessary, bridging streams and
valleys, making the crooked straight, and the rough places plain. And the
foundations of all the structures ever built by man, from the simplest sacrificial
altar of primitive man to launching sites for rockets aimed at the moon, depend
for their essential stability upon their contact with the earth.

As man's mind developed and a sense of inquiry was generated into problems
beyond that of daily subsistence, it was natural that the character of the ground
on which he walked and lived and had his being should engage his lively inter-
est. From such vague wonderings has developed the science of geology as it is
known today. So intimate would appear to be the links between the study of
the earth's crust and the daily activities of man that the inclusion of such a
paper as this in a commemorative volume might at first sight appear to be
superfluous. Geological studies might be thought always to be prosecuted in
the service of man. Indirectly, this is probably true, but in the recent history
of the science there appears to have been a tendency in some quarters to regard
geological inquiry as the end in itself; its application to the activities of man is
a slightly degrading use of pure scientific endeavour. A review of the way in
which geology has developed through the ages, with special reference to its
practical applications, may not therefore be out of place in this collection of

242

GEOLOGY IN THE SERVICE OF MAN 243

papers prepared in recognition of the seventy-fifth anniversary of a society
devoted to the promotion of geology in all its aspects. For the wheel appears
to have turned full circle.

The beginnings of geology' were rooted in the practical demands of early
man; the early days of the modern science led to the flowering of pure geological
study, much of it far removed from any thought of application; while today
the practice of modern engineering in many of its branches is making such
demands upon geology that there are now organized groups of petroleum and
engineering geologists. The challenge of the next few decades gives promise
of still wider use of geology in the service of man, concurrently with its advance
and development as one of the truly great branches of natural science.

The fact that "earth" was one of Aristotle's four elements is a clear indica-
tion of the importance with which the Greeks regarded the study of the ground.
Equally revealing are the frequent references to rock, stone, and soil in the
"History" of Herodotus. He knew personally the great stone quarries near
Memphis and "observed that there were shells upon the hills" during his stay
in Egypt. Was he the first palaeontologist? He might equally have been the
first to observe sulphate attack on structures, for he records that he saw "that
salt exuded from the soil to such an extent as even to injure the pyramids."
He certainly appreciated the engineering significance of geology*, for he makes
quite a few references to unusual excavation and river-diversion works carried
out by warlike monarchs. Cyrus and his diversion of the Euphrates provide
but one example. Two of the early queens of Babylon were similarly active,
Semiramis who constructed massive flood-control embankments of soil, and
the beautiful Nitrocris, who also diverted the Euphrates, but only temporarily,
while its natural channel through the city was lined with bricks of burnt clay
and a bridge of stone built over it "in the dry." (Herodotus, trans., 1910)

The contributions of Aristotle to early geological thought have often been
quoted. Even though it seems probable that his own book on rocks and min-
erals was lost, it is pleasant to imagine that it was Aristotle who inspired Alex-
ander to send out the special force that he directed to survey his empire, col-
lecting information on the "natural history" of the districts in which they were
at work, while maintaining the condition of the main roads. Anyone who has
had the privilege of following Alexander's route down the Kabul River from
what is now Afghanistan and then across the plains of Pakistan through Pesha-
war to Lahore can have no doubt that a lively appreciation of geology* must
have been yet another attribute of that commanding figure of history.

That these are not isolated cases from classical records of two thousand years
ago can be shown by reference to Vitruvius, a Roman architect and engineer,
of whom almost nothing is known except his famous "Ten Books on Architec-
ture," believed to have been written in the first century B.C. Some of his com-
ments on geological matters are almost uncanny in their modernity. He could

244 ROBERT F. LEGGET

\\ell be the patron saint of the geobotanists, for he relates (Book I, Chapter V)
that cattle on two sides of the river Pothereus, in Crete, were found to differ
in their spleens. Physicians found the explanation in an herb which grew on
one side of the river but not on the other. "From food and water, then, we may
learn whether sites are naturally unhealthy or healthy*" is his prophetic con-
clusion. (Yitruvius, trans., 1914)

His treatise includes a lengthy section on natural building materials; most
of his Book II is devoted to this subject. One would expect to find descriptions
of sand, lime, stone, and pozzolana, but who would expect to find in a book
written almost two thousand years ago the suggestion that "since the stones
are soft and porous, they are apt to suck the moisture out of the mortar and so
to dry it up" (a lesson not yet learned by all modern builders)? Somewhat less
scientific are the instructions that Vitruvius records for the construction of
dining room floors, "filled in with charcoal compactly trodden down, a mortar
mixed of gravel, lime, and ashes is spread on to a depth of half a foot (giving)
the look of a black pavement. Hence, at dinner parties, whatever is poured
out of cups, or spirted from the mouth, no sooner falls than it dries up, and the
servants who wait there do not catch cold from that kind of floor, although
they may go barefoot."

The "Ten Books" of Vitruvius not only reflected the high standard of Roman
engineering at the time he wrote, but almost certainly exercised considerable
influence upon its continued excellence. For several centuries after he wrote,
the works of Roman engineers played their part in forming the Europe that
is known today. Many miles of Roman roads and many examples of major
Roman structures still stand and serve despite the passage of the centuries.
The fact that there are no known examples of cyclopean or polygonal Roman
masonry, all examples of Roman building extant being of rectangular blocks
of stone set in regular courses, suggests that Roman engineers had a lively
appreciation of the importance of the proper selection of building stone and its
quarrying; their careful and selective use of travertine provides one specific
example. Their discovery of the unusual properties of the volcanic ash in the
vicinity of Naples, now called pozzolana but then pulvis Puteolanus, and espe-
cially the combination of this material with lime, opened up new prospects for
Roman building and led to such masterpieces as the Pantheon. And although
tunneling was not frequently an activity of the Romans, such tunnels as they
are known to have constructed demonstrate fairly clearly an appreciation of
the geology of the rocks penetrated. The use of vinegar poured over heated
limestone as a means of excavation, barbaric in its effect upon slave workers but
effective when used on the right type of rock, provides one unusual illustration.

It may be said that these examples are not of the applications of geology to
the works of man, but merely incidents in the development of the ordinary
practice of engineering; the name geology did not come into use for another

GEOLOGY IN THE SERVICE OF MAN 245

thousand years. But may not the same comment be made about the early
phases of all major branches of science and their applications? Chemistry
and physics did not spring full blown into the experience of man but started
haltingly through the probing inquiries of interested men; as a matter of fact,
the general acceptance of the terms "physics" and "chemistry" and their
recognition as distinct disciplines are relatively recent developments. One
would imagine that geology acquired such a lead over those other branches of
science, however, that it would have remained in the van of scientific develop-
ment throughout the centuries. This was not to be. The grandeur of Rome
departed. Human development entered its long centuries of delayed advance.
Recent historical studies are showing that the so-called Dark Ages were not
devoid of all progress, as is so often imagined, but the tempo of development
was sadly slowed throughout the Western world. There jvere advances in other
parts of the world, some of great significance, but for a thousand years or more
Western man seemed to be content with the physical world as he knew it, and
the minds that probed and the hands that experimented were few indeed.
When one takes a general look at the growth of scientific thought during this
long period, such as is so splendidly afforded by Charles Singer's eloquent
volume, "A Short History of Scientific Thought to 1900," the retarded position
of geological study in what little scientific advance did take place becomes all
the more puzzling.

The Arab world, even though it contributed so notably to mathematical
and astronomical knowledge, appears to have had very little interest in the
study of the earth. It seems probable that the attitude of the Christian Church
prior to the Reformation may have retarded geological studies. And yet as
one stands in one of the great mediaeval abbey churches, still in use today as
cathedrals in all their beauty, it is difficult to imagine that the learned monks
who built them accepted without question the varied stones that were quarried
for them and which they set in place with such obvious skill. To see in Wells
Cathedral in the west of England, for example, the great inverted arches be-
neath the central tower, installed as a preventive measure in the fourteenth
century, fills one with admiration not only for the skill in building they repre-
sent but also for the remarkable way in which the monks were able to obviate
a catastrophic foundation failure which they must have envisaged as they
developed their unique solution.

The records remain silent, however, and so the mystery of geological neglect
must stand. Even the renaissance of learning in western Europe failed to
change this strange stagnation of geological inquiry. Leonardo da Vinci quite
naturally applied his unique genius to the study of the earth, as to so much
else. His first observations on mountains were recorded in the year 1508 but,
again as in other fields, his acute observations failed to lead to any general
awakening of geological study. It was, indeed, not until the nineteenth century

246 ROBERT F. LEGGET

was well started that the first general stirrings of real interest in the earth's
crust and in its constituents are to be found. And yet, by this time, Laplace
had published his monumental "Celestial Mechanics,*' the most comprehensive
survey of astronomical knowledge ever made and a work that can be studied
with profit even today.

The contrast is indeed striking, so remarkable in fact that it demands an
explanation. Could it be that, unlike the case with these other sciences, the
awakening of modern geological study had to await the start of the industrial
age and the practical demands of modern engineering for information about
the earth's crust and the materials in it? At first sight, such a suggestion might
appear to be verging on the absurd, so subservient have been the applications
of geology to the prosecution of basic geological studies until comparatively
recent years. When, however, the records of the great pioneers of geology are
studied, some warrant for the suggestion becomes immediately apparent.

It is, for example, well known that the first of these modern "giants of
geology" (as they have been happily called), Abraham Gottlob Werner, was
appointed as an inspector of mines and instructor in mining and mineralogy
at the Freiburg Academy in 1775 when he was only twenty-five years old.
Here he conducted his famous classes for almost forty years until the disturb-
ances in the wake of Napoleon's army of occupation made it impossible for
him to continue. His skill as a teacher has never been questioned, even though
some of the theories he taught became the centre of great controversy. His
teaching, however, was specifically related to the application of geology in
mining so that he was essentially an applied geologist. Even his antipathy to
writing and his procrastination in relation to what little he did write fore-
shadowed all too clearly what has (unfortunately) become almost a character-
istic of those concerned with the application of the science of geology in the
service of man. (Fenton, 1952)

It is, however, not so generally recognized that Werner's great Scottish
antagonist, James Hutton, came to enjoy his interest in geology by way of a
very practical route; his importance in the general picture of unfolding geo-
logical theory obscured his earlier interests in applied geology. After his early
training in medicine and his disappointment at the prospects he saw in the
medical profession, he turned to his friend James Davie and shared an interest
with him in the production of sal ammoniac, while at the same time he farmed
the small estate left to him by his father. To this latter task he applied his
unusual scientific acumen, traveling extensively in Europe to study the most
advanced farming methods he could find. His farm became a show place;
and it is clear beyond all reasonable doubt that his farming activity introduced
Hutton to the scientific study of the earth, for he certainly noticed the relation
of soil properties to the character of the underlying rocks, as well as the insidious
effects of soil erosion. \Vhen he rented his farm in 1768 and moved to Edin-

GEOLOGY IN THE SERVICE OF MAN 247

burgh to devote himself to his scientific studies, he was forty-two years old,
with much valuable experience behind him and his geological interests fully
awakened through his practical activities and particularly his scientific farming.

D'Aubuisson and von Buch, having been students of Werner, might be
expected to have had interest in the applications of geology in mining. Von
Buch actually spent a year as inspector of mines in Silesia before coming into
the inheritance which gave him the freedom to pursue his scientific interests
as he wished. It is well known that when, in 1818, Adam Sedgwick was elected
to the chair of geology at Cambridge, he knew little of the science, his sound
classical education having apparently won his election. But he was very soon
learning his geology in the field, and he started in British lead mines, proceed-
ing to study the copper mines of Staffordshire and then the great salt mines at
Northwich in Cheshire. In his later years, even Sir Charles Lyell did not
disdain to study accidents in mines, despite the many demands upon his time
and his extensive travels.

Of even more significance is the fact that in his presidential address to the
Geological Society of London in February 1836, Lyell, after announcing the
award of the Wollaston Medal to Agassiz, went on to record his own part in
the formation of the Geological Survey of England, the first national geological
survey to be formed in the world:

Early in the spring of last year application was made by the Master General
and Board of Ordnance to Dr. Buckland and Mr. Sedgwick, as Professors
of Geology in the Universities of Oxford and Cambridge, and to myself, as
President of this Society, to offer our opinion as to the expediency of com-
bining a geological examination of the English counties with the geographical
survey now in progress. In compliance with this requisition we drew up a
joint report, in which we endeavoured to state fully our opinion as to the
great advantages which must accrue from such an undertaking not only as
calculated to promote geological science, which would alone be a sufficient
object, but also as a work of great practical utility, bearing on agriculture,
mining, road-making, the formation of canals and railroads, and other
branches of national industry. (Lyell, 1836, p. 358)

It is, perhaps, of even more significance to find a parallel development in
the start of modern civil engineering. The great British pioneer in this field
was originally a Yorkshire stone mason. John Smeaton (1724-1792) developed
such skill in the design and construction of notable structures that he is widely
regarded as the first civil engineer. His masterpiece was the first stone Eddy-
stone Lighthouse, the remains of which may still be seen standing on Plymouth
Hoe. His stone tower was built between 1756 and 1759. Many features of its
construction are noteworthy, such as Smeaton's research into the Roman use
of pozzolana and his use of the same material. Fortunately, Smeaton wrote
an account of the entire project and the geological insights that this record

248 ROBERT F. LEGGET

contains are revealing indeed. His own training as a stone mason could be
held to account for the detail in which he describes his search for the correct
type of stone of which to build the lighthouse, but the way in which he exam-
ined the treacherous rock on which he had to build the structure goes far beyond
what a mason might have been expected to do, especially in such an early and
pioneer venture. Smeaton's own words are worthy of brief quotation:

The congeries of rocks called Edystone appear to me to be all of the same
kind of stone, and of a kind so peculiar that I have not seen any stone exactly
like it in Cornwall or Devonshire ... It differs from the Moorstone in this;
instead of being composed of grains or small fragments, united by a strong
cement, interspersed with a shining talky substance, as the Cornish moor-
stone appears to be; it is composed of the like matter formed into laminae
commonly from one twentieth to one sixth part of an inch in thickness . . .
as is nearly one foot dip to the westward, in two feet horizontal, that is, in
an angle of about 26 degrees with the horizon. (Smeaton, 1791, p. 12)

These are the words of a practical civil engineer writing as early as 1791.

The nineteenth century had opened before engineering work really began to
transform the physical face of Europe, but the appreciation of geology so well
demonstrated by John Smeaton was shared by other leaders in the newly
developing profession of civil engineering. Canal and road building created
much building activity in Great Britain.

James Loudon McAdam was one of the towering practical road builders of
these early days. He, too, published a record of his ideas on road building, a
book that went through nine editions. A keenly developed sense of geological
appreciation permeates the book, such as in this statement about road-building
materials:

Flint makes an excellent road, if due attention be paid to the size, but from
want of attention, many of the flint roads are rough, loose and expensive.
Limestone when properly prepared and applied makes a smooth solid road
and becomes consolidated sooner than any other material; but from its
nature it is not the most lasting. Whinstone is the most durable of all ma-
terials; and wherever it is well and judiciously applied, the roads are compara-
tively good and cheap. (McAdam, 1823, p. 20)

Thomas Telford, undoubtedly the greatest of these early British civil engi-
neers, a man whose fame led to his being invited to undertake work in Sweden
and other European countries, was a man of action rather than words. For-
tunately, in addition to his own brief writings, we have records of his character
and works from other writers, amongst them Robert Southey, the English poet.
He accompanied Telford on an inspection tour, in 1819, through the Scottish
Highlands, where Telford was responsible for many projects including the

GEOLOGY IN THE SERVICE OF MAN 249

Caledonian Canal. Southey kept a journal and this has been published pri-
vately by the Institution of Civil Engineers. To find regular geological refer-
ences in a journal kept by the Poet Laureate of England suggests that they
must have been a regular part of the conversation of the two friends as Telford,
the engineer, explained to Southey, the poet, details of the many works they
had examined together. Southey noted that a new pier at Bervie Harbour
was being constructed of "pudding stone ... of all stone it is the worst for
working; but it is hard and durable, and when in place will do as well as if it
were granite or marble." They saw a new bridge at Forres "built of granite
of all colours . . . (near to which is) a bank of granite, in such a state of decom-
position, or imperfect composition, that it crumbles at a touch." Southey tells
how Telford advised a laird in Skye about developing a marble deposit, saving
him from great expenditure by persuading him to make first an experimental
shipment to London. And in a deposit of "marie" near Inverness they found
it to be "full of very small shells, some resembling whelks, others like the fresh
water muscle, but all very small." (Southey, 1929)

The position of William Smith (1769-1839) can now be seen to be no iso-
lated phenomenon, as is sometimes suggested, but rather the supreme example
of what was probably almost commonplace in the early years of civil engineer-
ing. For William Smith was a civil engineer; he so signed his name in docu-
ments still to be seen. It is almost amusing to note attempts in some geological
writing to skirt around this basic fact, by such devices as referring to Smith
as a surveyor or a land agent. Fortunately, the engineering work of the man
who so well earned the title of "Father of British Geology" is well documented
and it can be seen to be as thorough as were his geological studies, upon which
his fame so firmly rests. His whole professional career was so intimate a blend-
ing of geology and its application to the practice of engineering that it is
hardly necessary to do more than make this brief reference to it. But it may be
useful to recall that his famous map appeared (in 1815) bearing a long title
which began "A Delineation of the Strata of England and Wales, with part of
Scotland, exhibiting the Collieries and Mines ..." There was no doubt as to
William Smith's appreciation of the value of applied geology, nor of the breadth
of his vision as to how geology may be applied in the service of man, for (in
his own words) his "New Geological Adas of England and Wales," which
followed the famous map between 1819 and 1824, was "calculated to elucidate
the Agriculture of each County, and to show the Situation of the best material
for Building, Making of Roads, the Construction of Canals, and pointing out
those Places where Coal and other Valuable Materials are likely to be found."
This must surely be one of the earliest charters of applied geology. (Phillips,
1844)

What of geology in the new world? Its indebtedness to the early geologists
of Europe is well appreciated, but it may be asked whether the start of the

250 ROBERT F. LEGGET

scientific study of geology- in North America had this same practical beginning.
The answer is clear as soon as the records of early American geologists are
consulted. William Maclure started his geological studies with the objective
of assisting with the development of ores of iron, copper and zinc. His last
years were taken up in part with the start of his school of agriculture because
the scientific study of soil was one of his continuing interests. James Hall's
early work in New York State resulted in a report (1843) that devoted much
attention to mining matters as well as giving the first account of the geological
sequence to be found in that state. (Fenton, 1952)

In early Canada, with its widely scattered and sparse population, the same
picture is found, epitomized in another commanding founder of modern
geology, Sir William Ernest Logan, first Director of the Geological Survey of
Canada. Born in Montreal in 1~98, Logan was sent to school in Edinburgh
and then entered the business house of an uncle in London. In 1831 he moved
to Wales to take charge of a copper smelter. He became interested in the local
coal mining and prepared a geological map of the coal field, without any
instruction. This led him to a detailed study of the underclays associated with
coal, and he presented a paper about them to the Geological Society (of
London) in 1840. He sailed for Halifax in 1841 and immediately upon landing
started on those field studies of Canadian geology' the record of which makes
exciting reading even today, enlivened as it is by his quite brilliant pen and ink
sketches.

His work soon attracted attention, with the result that in 1842 he was ap-
pointed director of a Geological Survey 15 years before the Dominion of
Canada itself had been constituted. In supporting his candidature, Sir Henry
de la Beche, Director of the British Survey, stated that "I would further observe
that Mr. Logan is highly qualified as a miner and metallurgist to point out the
applications of geology- to the useful purposes of life, an object of the highest
importance in a country like Canada, the mineral wealth of which is now so
little known." 1 From the time of his appointment until his eventual retirement
in 1869 (he died in 1875J, he seemed to do the work of several men, becoming
almost a legendary figure throughout eastern Canada.

The practical aspect of his work was evident from the use to which his in-
vestigations were put by such great engineers as Thomas Keefer and George
Stephenson (for the design of the piers of the Victoria Bridge, Montreal, piers
that are still in use). He received many offers of other, and more gainful,
employment. In writing about one such offer, he had this interesting observa-
tion to make.

When the British Government gave up the Michigan territory at the end
of the last American war, with as little concern as if it had been so much
bare granite, I dare say they were not aware that 12,000 square miles of a
coal-field existed in the heart of it larger than the largest in Britain, though
the smallest of those belonging to the United States, which possess another

GEOLOGY IN THE SERVICE OF MAN 251

of 55,000 square miles, and a third of 60,000 square miles . . . Taking all
this into consideration, notwithstanding I have requested my brother Ed-
mond, of Edinburgh, who has a friend in the East India direction, to make
some inquiry into the matter, I fancy you will see that the chances are that
I am tied to Canada. (Harrington, 1883, pp. 235-236)

So germane to this study is Logan's whole philosophy that the temptation
to quote from his own lucid writings is hard to resist. One more citation may
perhaps be admitted, his own definition of the work of the Geological Survey
of Canada.

The object of the Survey is to ascertain the mineral resources of the coun-
try, and this is kept steadily in view. Whatever new scientific facts have
resulted from it, have come out in the course of what I conceive to be eco-
nomic researches carried on in a scientific way . . . The analyses of new
mineral species, while they directly regard a scientific result, must always
have an economic bearing. You cannot tell whether a new substance is to
be profitably available or not until you have ascertained its properties. The
analyses of mineral species led to our knowledge of the limefeldspars, of so
much agricultural importance to the Laurentian country. Thus economics
lead to science, and science to economics. (Harrington, 1883, pp. 293-294)

Economics lead to science, and science to economics and this from the
man whose studies of the Precambrian have stood the test of a century of
further study.

Words of Edmund Burke come vividly to mind as these glimpses at the early
days of geological study are considered: "People will not look forward to pos-
terity who never look backward to their ancestors." A seventy-fifth anniversary
is not only a time for glancing backward, not alone a time for reviewing present
achievements, but also a time for looking ahead to the challenges of the future.
How is geology to serve mankind in the years ahead as it so surely did in the
days of its beginnings?

The years between have a natural bearing upon the answer to be given, even
though the record they present, certainly for the remainder of the nineteenth
century, is not inspiring. Despite such auspicious beginnings, the divorce of
geology from all its applications became very real as the century advanced, to
such an extent that, in 1880, an eminent British geologist is reported to have
said that the amount of money fruitlessly spent in Great Britain in a ridiculous
search for coal, even within his own memory, would have paid the entire cost
of the British Geological Survey. Early mining records are replete with
examples of geological neglect, all supporting statements in this vein.

Developments in the application of geology in civil engineering followed a
similar course. The fault almost certainly must be shared equally by geologists

252 ROBERT F. LEGGET

and engineers, the mutual respect of early years giving way to differences that
were sometimes acrimonious. Another of the great pioneer British engineers
was involved in such divergences which, although now perhaps amusing, must
have had singularly unfortunate results at the time.

Isambard Kingdom Brunei was a young man of genius who gained his
experience from his distinguished father, Mark Isambard Brunei (born in
France), who was at one time Chief Engineer of the City of New York. In 1824
Mark Brunei was responsible for the start of work on the first subaqueous tunnel,
beneath the River Thames at Rotherhithe, just east of London. On the basis
of numerous trial borings, geologists assured the older Brunei that he would
find a stratum of strong blue clay and avoid the quicksand that had plagued
an earlier attempt at tunneling. The prediction proved to be quite wrong, and
although the tunnel was eventually finished by the elder Brunei, the extra cost,
incredible difficulties, and the hazards that had to be overcome must have left
an indelible imprint on the young man's mind.

When building the Great Western Railway, almost twenty years later, the
younger Brunei faced another major tunneling job in the construction of the
two-mile Box Tunnel between Chippenham and Bath. The tunnel penetrates
blue clay, blue marl, the Inferior Oolite and the Great Oolite, or Bath Stone.
After the tunnel had been completed and put into use the Reverend Doctor
William Buckland, then at Oxford, caused much trouble by declaring, even
though he had not visited the tunnel, that the unlined section was highly
dangerous and would certainly fall "owing to the concussion of the atmosphere
and the vibration caused by the trains." To this Isambard Brunei replied that
although he regretted his lack of scientific knowledge of geology, he had had
experience in excavating the rock in question and so considered it to be quite
safe. A Board of Trade inspector was called upon to report on the safety of
the tunnel and strongly supported Brunei. Some of the tunnel remains unlined
to this day, 120 years after its opening. (Rolt, 1961)

Pursuit of such arguments, disagreements, and outright mistakes, although
possibly of some human interest, is not profitable in a study such as this. The
schism was created, however, and persisted throughout the century. There
were naturally individual exceptions and occasional glimpses of the benefits
to be derived by cooperation, such as a series of articles by W. H. Penning
published in the British journal The Engineer in 1879, and reprinted a year
later in book form as "Engineering Geology." (Penning, 1880)

The academic isolation of the science almost certainly led eventually to its
neglect in the sphere of public interest, represented so unfortunately in its
almost complete absence from the curricula of schools, a neglect that persists
even today, and its general denigration until very recently in the activities of
natural history clubs and similar amateur scientific endeavours. This situation
is also in contrast with the position in earlier days. A rare book came recently

GEOLOGY IN THE SERVICE OF MAN 253

into the possession of the writer showing that almost a century ago no less a
man than the Reverend Charles Kingsley, author of "The Water Babies" and
other famous books for young people, was a popular lecturer on geology in
addition to his many other activities. The book, "Town Geology," contains
the text of lectures delivered in 1872 to the "young men of the city of Chester
(England)," presumably at meetings sponsored by the Chester Natural History
Society. In his introduction, Kingsley says:

It does seem to me strange, to use the mildest word, that people whose
destiny it is to live, even for a few short years, on this planet which we call
the earth, and who do not at all intend to live on it as hermits . . . shall in
general be so careless about the constitution of this same planet, and of the
laws and facts on which depend, not merely their comfort and their wealth,
but their health and their very lives, and the health and the lives of their
children and descendants. (Kingsley, 1877, pp. xv-xvi)

In rolling, somewhat ponderous but still enjoyable Victorian prose, the
author gives a lucid general picture of physical geology and then shows how it
affects "The Soil of the Field," "The Pebbles in the Streets," "The Slates on
the Roof," and similar very practical subjects. The science of geology, says
Kingsley, "is (or ought to be), in popular parlance, the people's science the
science by studying which, the man ignorant of Latin, Greek . . . can yet
become ... a truly scientific man." And in urging his readers to use their eyes,
even in their own town, he says: "Be sure, that wherever there is a river, even
a drain; and a stone quarry, or even a roadside bank; much more where there
is a sea, or a tidal aestuary, there is geology enough to be learnt, to explain the
greater part of the making of all the continents on the globe." (Kingsley,
1877, pp. 4, 35)

These words written so long ago are still strangely relevant. After decades
of popular neglect, geology is again beginning to win some degree of general
recognition; its application at last is appreciated by the man in the street as
something that affects him. University extension lectures are the modern
equivalent of such popular lecture series as those of Charles Kingsley. To these
have been added, however, the steady stream, flood it might almost be called,
of cheap popular literature, and all that modern mass communication media
can provide in purveying information. To see well-illustrated "paperbacks"
on geology on even small bookstalls is now commonplace; the same cannot
be said of other major scientific disciplines with the exception of meteorology.
When radio and television are used as educational media, geology is not neg-
lected. For example, the Canadian Broadcasting Corporation devoted one of
its major nationwide evening "University" lecture series to an introduction
to geology; later the lectures were reprinted in book form. (Baird, 1959)
Slowly, geology is at last winning a place for itself in the curricula of some high

254 ROBERT F. LEGGET

schools, and some organization such as the Bov Scouts, are giving it increased
recognition in their excellent training work. Once again, it is coming to be
recognized as "the people's science."

What has prompted this awakening of the public interest? The reality of
the change in attitude lends validity to the question. As in so many other social
changes, the automobile would seem to hold the key to at least a partial answer.
The potential mobility of the general public through the wide availability of
private automobiles, when coupled with the unaccustomed leisure and extra
wealth that the affluent society of North America is providing, has resulted in
an amount of traveling for the ordinary citizen such as the world has never
previously witnessed. And as the citizen travels, he must be conscious, to a
degree, of the engineering of the roads on which he travels, especially when
new highway construction interferes with his headlong high-speed touring.
He can then see, as he waits impatiently to pass in single line through some
construciion project, that civil engineers have to excavate rock and soil for
their cuttings and tunnels; some motorists at least must see that all rock is not
the same. Even while traveling, the motorist will be at least conscious of the
scenery through which he passes. Interest has been aroused to the degree that
there are available today many route guides giving geological and historical
information about unusually interesting routes. (W. Texas Geol. Soc., 1958)

Doubtless there are other causes of the regeneration of public interest in
geology, but reflection suggests that it is the work of the civil engineer, as seen
by the itinerant public, that is chiefly responsible. Few people have occasion
to visit mines; fewer still probably give thought to the source of the gasoline
they use so profligately; and, although adventures with mining stocks may
make denizens of the stock market familiar with the names of mines, real and
imaginary, this is no warranty of knowledge of geology- in fact, sober recita-
tion of geological facts would remove much of the adventure from the bucket
shops. As a prime example of the application of geology in the service of man,
the practice of present day "engineering geology," as it has come to be called,
will therefore be examined in brief review against the broader context of the
unfolding science itself before a final glance is taken at what the future
may hold. If, as William James once remarked, philosophy in the full
sense is only man thinking, such a review might suggest a philosophy of
applied geology.

Just as the work of William Smith epitomizes some of the earliest interrela-
tions of geology and civil engineering, so the work of Charles P. Berkey marks
a real turning point in the more recent history of the application of geology
to civil engineering work. In 1905, he was retained in connection with the
construction of the Catskill Aqueduct for the water supply of the City of New
York. Thereafter for half a century, his services were hi constant demand in
association with many of the greatest civil engineering projects of North Amer-

GEOLOGY IN THE SERVICE OF MAN 255

ica. Dr. Berkey's outstanding contribution to the development of the Geo-
logical Society of America makes this reference to his work singularly appropri-
ate. After serving as Secretary to the Society for eighteen years, he was elected
President in 1941 and in the same year became an Honorary Member of the
American Society of Civil Engineers, one of the many honours bestowed upon
him which he always valued in a rather special way. In his presidential address
Dr. Berkey had this to say about the assistance that geologists can render to
civil engineers in the prosecution of:

. . . works marking the progress of material civilization and magnifying the
limited competence of man's bare hands. They are literally fountains of
power perhaps in this present emergency the power that is to save the
world. In a world seemingly bent on self-destruction, at least these adven-
turers face toward the future and stand out as if to promise better times in
a more constructive world. So I claim a place of honor for these men who
spent their lives in devising new ways of using their specialistic knowledge
and experience and ingenuity for more effective public works and for the
greater comfort and safety of men and women everywhere. (Berkey, 1942,
p. 531)

These words were spoken twenty years ago; how relevant they are to the world
of today.

It was in 1928, at the age of 61, that Berkey was retained by the U. S. Bureau
of Reclamation just as the monumental Hoover Dam reached the final stages
of planning. His assistance with the study of the foundation problems and the
allied diversion tunnels associated with this great structure was recorded in a
series of lucid reports that in no way minimize the details of the relevant
geology but which make clear how the safety of the engineering structure
depends ultimately and entirely upon the adequacy of the supporting rock
strata. This point of view is now very generally accepted, to such an extent
that the failure of a dam because of geological defects in its foundations is
today almost unknown. The tragedy of the failure of the Malpasset Dam in
1959, and the exhaustive investigations into the cause of the disaster serve to
confirm the foregoing suggestion in no uncertain manner.

All the tools and methods of geological investigation geophysical methods,
the use of large and small bore holes, exploratory shafts and tunnels all supple-
mentary to detailed geological surveying are now regularly used not only for
the study of dam sites but also for determining the geological conditions to be
anticipated along the routes of proposed tunnels. In studies for the long-
discussed Channel Tunnel between England and France, comparative studies
of microfossils (similar to the widely followed practice in petroleum engineering)
were carried out in some detail in checking the assumed continuity of the
Lower Chalk across the bed of the Strait of Dover. (Bruckshaw, et al., 1961)

256 ROBERT F. LEGGET

In all applications of geology to the works of the civil engineer there are
reciprocal benefits of varying degree but the excavation of tunnels provides
perhaps the most obvious examples. The civil engineer can be guided in his
work by the preliminary predictions of the geologist; the geologist can obtain
invaluable information if he is permitted to examine the rocks exposed as the
tunnel advances. There is on record the particularly delightful incident which
permitted Dr. Hans Cloos finally to solve the riddle of the Rhine Graben by
his observations in the short tunnel beneath the Lorettoberg at Freiburg.
(Cloos, 1954) New light was shed on some of the complexities of the geology
of the Scottish Highlands by observations made as the great Lochaber water-
power tunnel was excavated beneath the slopes of Ben Nevis. (Peach, 1930)
And in more recent years, one of the most difficult tunnels ever driven, the
Tecolote Tunnel in California, was put to good geological use before being
placed in service; it was used for a detailed study of the trends in the Tertiary
strata that it penetrates. (Bandy and Kolpack, 1962) The dewatering, by
means of cofferdams, of dam sites provides correspondingly unique opportuni-
ties for the actual study of riverbed geology.

The construction of major bridges today always includes a detailed study of
one or more river or lake crossings and the requisite test borings will sometimes
reveal submarine geological features that were previously unsuspected. Dis-
covery of a preglacial gorge in Mackinac Strait between lakes Michigan and
Huron in the course of foundation studies for the great suspension bridge which
now spans the Strait is a good and quite typical example. (Rosenau, 1958)
Civil engineering structures built in the open sea necessitate unusually difficult
site investigations which will always provide a dividend in the way of new
information on submarine geology. Of unusual value has been the information
on the deltaic deposits of the Mississippi River derived from soil mechanics
studies of soil samples obtained in connection with site studies for offshore oil
drilling platforms. (Fisk & McClelland, 1959)

Even more mundane engineering works such as the construction of roads
and airports have their own contribution to make to the fund of geological
knowledge, especially by the new sections revealed by major excavation work.
At the same time, the wide areal coverage of such projects has led to develop-
ments in methods of field investigation which constitute advances in geological
technique. Notable has been the use of geophysical methods (especially seismic)
for the study of shallow deposits such as those in glaciated areas. Engineering
use of geophysical methods for subsurface investigation has constituted a singu-
larly useful link between geology and geophysics, a link that must be greatly
strengthened in the years ahead as the concept of quantitative measurement
becomes more widespread in geological practice.

Geological conditions on quite restricted building sites, such as those in
central urban areas, are now engaging the attention of engineers to a degree

GEOLOGY IN THE SERVICE OF MAN 257

never experienced previously. The increasing complexity of foundation con-
struction in crowded city areas, due to the proliferation of subsurface services,
makes knowledge of exact site conditions a virtual necessity. As geological
records of this very localized type are assembled and coordinated, valuable
information about the geology beneath city streets is developed. Urban geology
has therefore become a term of real significance and is not the mutually con-
tradictory juxtaposition of words that it might at first sight appear to be.

There are some critical urban areas where a combination of steeply sloping
ground and unsatisfactory soil conditions gives rise to serious landslide hazards,
particularly after periods of heavy rain. Some of the suburbs around the city
of St. Louis, Missouri, are plagued by this danger, but the largest and most
notable area to have been developed in the face of this hazard is that constituted
by the city and county of Los Angeles. So serious has this local situation become
that legislation has been enacted requiring a report on any potentially unstable
site by an engineering geologist before development work can proceed. Here
indeed is a very obvious example of geology in the service of man. (Grove,
1957) This legal requirement has led to the establishment of a professional
society, the California Association of Engineering Geologists.

Antedating this group by more than a decade is the Division of Engineering
Geology of the Geological Society of America. The establishment of this divi-
sion (in 1947) was public recognition, by the Society, of the importance of this
particular application of geology. Both of these organizations, however, are
fledglings in comparison with the geological divisions of several national insti-
tutions of mining engineers, whose continued work has long provided valuable
records of mining activity. Not much younger than the Geological Society
itself are two other North American societies serving the needs of applied
geology. Both are now international in scope, and their example has now been
followed at the local level by the establishment of more than a few local mining
or petroleum geological societies.

The formation of the Society of Economic Geologists in 1920 and of the
American Association of Petroleum Geologists in their more specialized field,
in 1917, provide further turning points in the applications of geology. It is
significant that alternatives considered to the name finally adopted for the
Society of Economic Geologists included the Society of Geological Engineers
and the Society of Applied Geology. Notable also is the fact that at the Society's
first meeting there was presented a paper on "The Relation of Economic
Geology to the General Principles of Geology" by President R. A. F. Penrose,
Jr. (1921).

It is not often that the official bulletin of a society antedates the society
itself, but this is the case with the distinguished journal, Economic Geology,
which is the bulletin of the Society of Economic Geologists. The first issue
appeared in October 1905, and the Fiftieth Anniversary Volume in 1955. The

258 ROBERT F. LEGGET

two parts of this notable compilation, containing 1130 pages, present a compre-
hensive survey of all main aspects of the geology of ore deposits, to which the
journal itself has been generally devoted, but review papers dealing with engi-
neering geology, soil mechanics, and ground water are also included. It is
impossible to make even a cursory examination of this significant collection
of papers without appreciating how much the application of geology in the
search for ores and fuel has contributed to significant advances in the science
of geology itself.

The start of exploration for petroleum in Oklahoma about 1913 led to great
interest in this area in the application of geology to the discovery and winning
of oil. At a dinner in Tulsa, Oklahoma, on October 2, 1915, it was decided
to form an association of geologists interested in this relatively new aspect of
applied geology*. The Southwestern Association of Petroleum Geologists was
organized on February 19, 1917, also in Tulsa, but the name was changed in
the following year to the American Association of Petroleum Geologists. The
Association now has a worldwide membership of more than 1 5,000 a number
that is clearly indicative of the importance of this branch of applied geology.
Through its Bulletin, associated specialized journals, and its more than thirty
special volumes, the Association has made significant contributions to the litera-
ture of geology. As is the case with Economic Geology, the Bulletin of A.A.P.G.
includes many papers that may properly be described as theoretical, some seem-
ingly far removed from the search for petroleum.

The search for minerals and fuels will long continue. Exploration for new
reserves of petroleum is today being pursued throughout the world on an un-
precedented scale, to such an extent that it undoubtedly represents at present
the most extensive application of geology in the service of man. Inevitably,
however, it must be a transitory application even in comparison with the cor-
responding search for minerals. All who are familiar with the Paley report
and what geologist is not? will know the challenge that even now exists in
the mineral field. And all such exploratory work will involve the most assiduous
application of the principles and techniques of geology- and geophysics, if
indeed the two need to be separated; it will involve deep probings beneath the
surface of the earth and the bottom of the sea; and in this and many other
ways, it will add to the store of geological knowledge and doubtless make its
own special contributions to the advance of geological theory and principles.
Economics will long lean on science, and science upon economics.

As the future becomes the present, and the present the past, who can tell
what advances will be seen in this unceasing hunt for the treasures of the earth?
It is when one considers carefully the surface of the earth that the most impor-
tant of all the possible applications of geology in the service of man comes into
view. The population of the world is now exploding, and the rate of increase
still shows no significant signs of slackening. Even assuming that some of the

GEOLOGY IN THE SERVICE OF MAN 259

worldwide measures for population control that can now be seen to be impera-
tive are introduced before it is too late, it is safe to say that by the year 2000 A.D.
a crucial point in time that will be seen by most readers of this volume the
population of the world will be almost double what it is today. To maintain
these more than 6 billion lives, the use of almost all available arable land will
be necessary as will also supplies of water that will make public water-supply
systems of the present day appear almost parsimonious in comparison. Con-
servation of the renewable resources of the earth, notably soil and water, is
therefore a public responsibility of immediate urgency.

Some may be tempted to say what has this to do with geology? Just this,
that in the conservation of the renewable resources of the earth for man's
future use, and especially water, civil engineering works will be called for on a
scale surpassing anything yet seen, except in certain carefully developed river
valleys. And in all these works the aid of the geologist will be essential, particu-
larly in all considerations of subsurface water and of soil-erosion control.
These projects will be in addition to the works and buildings required for the
ordinary service of communities and areas, in the execution of which close
collaboration of engineer and geologist will be vitally necessary. In all this
work the engineer must have essential geological information for the stability
and safety of his structures, and the geologist will gain new insights into sub-
surface conditions as these are revealed by the excavations of the engineer.
In this way the bounds of geological knowledge will steadily increase.

What, then, is the conclusion of the matter? Even in such a pedestrian review
as this, it has not been possible even to mention all the main applications of
geology in the service of man. Military geology, for example, is now a well-
accepted branch of applied geology upon which many comments could have
been made. Much more could have been said about the new techniques that
even now can be seen to have great value in all such engineering projects. It
is exceedingly clear, however, that geology will be applied in the works of the
engineer in steadily increasing measure as far into the future as the mind can
foresee.

It has been well said that merely boring holes in the ground is not geology.
But if, in the location of test bore holes, geological advice is always sought and
in the interpretation of test bores geological significance is always considered,
the needs of the engineer can be better met while new information will thus
become available for the use of the geologist. It is clear that there is no special
philosophy of applied geology. How could there be in the application of the
science to the art of engineering? If, however, the geologist brings to bear in
his part of such cooperative endeavour a lively appreciation of the philosophical
aspects of his work, he will be the more appreciative of the philosophical back-
ground of all true engineering. And with such a meeting of the minds in the
prosecution of what must always be essentially detailed and utilitarian work,

260 ROBERT F. LEGGET

what potential is presented for advancing the boundaries of knowledge while
serving the needs of man !

Some must have thought it a hard saying as they were reading these words
of the late Mr. Justice Holmes: "Science teaches us a great deal about things
that are not really important, philosophy a very little about those that are
supremely so." This Olympian view might have been expected from the
eminent jurist. To the geologist at work in the field or in the laboratory, as to
the engineer engaged in the design or the construction of some vitally needed
public project, the words will seem to be almost heresy. And yet, as one sets
the acknowledged wonders of geological investigation or the applications of
geology' in even the greatest of engineering works against the backdrop of human
history and as one glimpses ahead into ages yet unborn, the words persist in
their challenge, reminding one of the dominance of the mind of man in all
such studies, of the human core of all understanding. This is no new thought,
for down through almost four centuries of time come words with which this
essay may most fitly close, words of a philosopher-scientist whose prescience so
frequently surprises his modern readers. It must have been with thoughts such
as those now ventured that Sir Francis Bacon recorded his great concern "To
make the mind of man, by help of art, a match for the nature of things."

REFERENCES CITED

BAIRD, D. M., 1959, An introduction to geology: Toronto, Canadian Broadcasting

Corp., Ill pp.
BANDY, O. L., and KOLPACK, R. L., 1962, Formaniferal and sedimentological trends in

the Tertiary section of Tecolote Tunnel, California: Geol. Soc. Am., Abs., Ann. Mtg.,

Houston, p. 8A.
BERKEY, C. P., 1942, The geologist in public works: Geol. Soc. Am., B., vol. 53, pp.

513-532.
BRUCKSHAW, J. M., GOOUEL, J., HARDING, H. J. B., and MALCOR, R., 1961, The work

of the Channel Tunnel Group 1958-1960: Inst. Civil Eng., Pr., vol. 18, pp. 149-178;

See also discussion in vol. 21, pp. 611-628.
CLOOS, H., 1954, Conversation with the earth (trans, by E. B. Garside): London,

Routledge and Kegan Paul, 402 pp.
FENTON, C. L. and M. A., 1952, Giants of geology: Garden City, N. Y., Doubleday,

333 PP .
FBK, H. N., and MCCLELLAND, B., 1959, Geology of continental shelf off Louisiana; its

influence on offshore foundation design: Geol. Soc. Am., B., vol. 70, pp. 1369-1394.
GROVE, D. E., 1957, A geologic challenge: GeoTimes (Am. Geol. Inst.), vol. 2, no. 5,

p. 8.
HARRINGTON, B. J., 1883, Life of Sir William E. Logan, Kt.: Montreal, Dawson Bros.,

432pp.

GEOLOGY IN THE SERVICE OF MAN 261

HERODOTUS, 1910, The History of Herodotus (trans, by G. Rawlinson), 2 vols.: London,

J. M. Dent and Sons, Ltd., 719 pp.

KINGSLEY, C., 1877, Town geology: London, Daldy, Isbister and Co., 239 pp.
LYELL, CHARLES, 1836, Geol. Soc. London, Pr., vol. 2, no. 44, pp. 357-361.
McADAM, J. L., 1827, Remarks on the present system of road making, 9th ed.: London,

Longman, Hurst, Rees, Orme, and Brown, 236 pp.
PEACH, B. N., 1930, The Lochaber water power scheme and its geological aspects:

Water and Water Eng. (London), vol. 32, p. 71.

PENNING, W. H., 1880, Engineering geology : London, Ballie>e, Tindall and Cox, 164 pp.
PENROSE, R. A. F., Jr., 1921, The relations of economic geology to the general principles

of geology: Econ. Geol., vol. 16, pp. 48-51.

PHILLIPS, J., 1844, Memoirs of William Smith, LL.D.: London, John Murray, 150 pp.
ROLT, L. T. C., 1961, Isambard Kingdom Brunei: London, Arrow Books, 383 pp.
ROSENAU, J. C., 1958, Geology of the Mackinac Bridge site, Michigan: Geol. Soc. Am.,

Abs., vol. 69, Ann. Mtg., St. Louis, p. 1636.
SINGER, C., 1959, A short history of scientific ideas to 1900: Oxford, Clarendon Press,

525 pp.
SMEATON, J., 1791, A narrative of the building and a description of the construction of

the Edystone lighthouse with stone . . .: London, H. Hughs, sold by G. Nicol, 198 pp.
SOUTHEY, R., 1929, Journal of a tour hi Scotland in 1819: London, John Murray, 276 pp.
VITRUVIUS, 1914, The Ten Books of Architecture (trans, by M. H. Morgan): Cambridge,

Harvard University Press, 331 pp.
West Texas Geological Society, 1958, Geological road log, Del Rio to El Paso: Midland,

West Texas Geol. Soc., 48 pp.

CLAUDE C. ALBRITTON, JR.

Southern Methodist University

Philosophy of Geology:

A Selected Bibliography
and Index 1

This bibliography contains references to some four hundred writings which
reflect upon the scope, methods, and contributions of the geological sciences.
The collection is not comprehensive; neither are its selections evenly balanced
with regard to their professional origins, geographic distribution, or years of
publication.

Because the main effort here is to identify live rather than fossil problems, I
have drawn more than half the items from works published since 1949. A geo-
graphical weighting in favor of American publications a little more than
half the entire list is unintentional. With more time, and with continued
access to competent translators of eastern European and Scandinavian lan-
guages, a better geographical balance could have been attained. I have made
a special effort to learn how geology is viewed by persons who are not geolo-
gists. Consequently, a fourth of the references are to the works of historians,
philosophers, mathematicians, physicists, astrophysicists, biologists, bibliog-
raphers, and linguists.

The index at the end of this paper tells something about degrees of interest
in different problems and subjects. As might be expected, three areas of spirited
discussion center around the concepts of time, change, and the classification of
natural phenomena.

Hutton, as one admiring author claims, discovered time. In any event, the
science he founded has seemed to differ from others by its concern with the
changes that have occurred over vast stretches of time. Because history is also
concerned with temporal sequences of events, geology has come to be known as

1 Prepared with the aid of National Science Foundation Grant G-23887.

262

A SELECTED BIBLIOGRAPHY AND INDEX 263

an historical science. This leads one to ask whether historians and historical
geologists use the same basic methods. No, say those historiographers who
insist that the only proper history is human history, and that in order to write
it, one must relive in the imagination the things that actually happened to
human beings in the past. Yes, say those philosophers who maintain that the
explanation of whatever has happened, immediatelv or remotely in the past,
requires the formulation and use of laws. However this particular issue may IDC
resolved, it would appear that historical geologists are called upon to explain
their explanations and to show \vhat use, if any, they make of laws.

"The present is the key to the past, 53 we say. The principle of uniformitv, to
which this adage refers, is held by many to be the foundation stone of geology.
With regard to the validity of the principle, however, the range of opinion is
amazing. Most modern historians of science seem to agree that LyelPs famous
principle was an a-historic device, which was discarded after evolutionism
became popular in the nineteenth century. Most modern philosophers of
science seem to feel either that the principle is too vague to be useful, or that
it is an unwarranted and unnecessary assumption. Geologists and other scien-
tists have such varied opinions on the matter that it would be impossible, with-
out a vote, to say which view prevails. Surely the principle of uniformity needs
the critical attention of geologists. Is it no more than a prescription for analogi-
cal reasoning? Will it reduce to the principle of simplicity, as applied to the
agencies of geological change? And if so, with which of the many varieties of
simplicity is it to be equated?

As Simpson has said, all theoretical science is ordering. Many of the papers
cited in the bibliography deal with the classification and ordering of natural
phenomena. Admittedly, the things studied by geologists and paleontologists
are hard to classify. Minerals, rocks, and natural fluids rarely occur in any
pure or ideal state. The surface of the earth is so irregular, both morphologically
and tectonically, that it is hard to devise classifications of land forms or geologi-
cal structures that will apply to all terranes. Stratigraphic units are rarely
clear cut, and it is difficult to name and group them in ways that will disclose
both their relationships in space and their sequences in time. In classifying
fossils, the paleontologist must contend with the imperfections of the fossil
record, as well as the variable influences of environmental change and evolu-
tionary development.

The greater part of this bibliography was assembled during the summer of
1962, which will explain why certain publications of more recent date are not
included. The search for titles began with a scanning of the indexes for all
issues of the "Bibliography and Index of North American Geology" (1732-
1959), and of the "Bibliography and Index of Geology Exclusive of North
America" (1933-1960). The search was continued through the files of "Phil-
osophy of Science," the "British Journal for the Philosophy of Science," and the

264 CLAUDE C. ALBRITTOX, JR.

''Zentralblatt fur Geologic und Palaontologie." "Isis," \\ith its annual bibli-
ographies for the history of science, yielded additional material. The bibli-
ographic card service of the Seiwce flnfot motion Geologiqiie du Bureau de Recheiches
Geobgiques et Mimbes led to many articles published in Russian journals since
1960.

I should like to thank the asjencv which made this work possible and the
individuals who helped brine it to an end A grant from the National Science
Foundation permitted me to give three months of concentrated effort to the
project. Mrs. Robert R. Wheeler did much of the searching for titles and
assembled the index. Mrs. Natalie Yoshinin also helped with the searching,
and she translated the Russian articles. To Mrs. Nadine George, Reference
Librarian of the Science Information Center in Dallas, I am much indebted
for her patient and fruitful efforts at locating out-of-the-way materials.

Many friends have called my attention to pertinent references they had run
across in their own reading. Dr. Arthur F. Hagner graciously sent me a lengthy
bibliographv of the philosophy of geologv, which he had prepared for his
advanced students at the University of Illinois. Miss Marie Siegrist has called
my attention to references which are to appear in future issues of the "Bibliog-
raphv of Geology Exclusive of North America.'' Mr. Charles Gotschalk, Chief
of the Stack and Reader Division of the Library of Congress, provided a private
room for my use during a month of study in Washington. Mrs. Jacquelyn
Ne\\ bury typed the manuscript and helped with the proofreading. Although
I am deeply grateful to all these persons, I must say, for their protection, that
I am solely responsible for the selection and annotation of titles, the design of
the index, and for any errors.

Bibliography

American Commission on Stratigraphic Nomenclature, 1961, Code of strati-
graphic nomenclature: Am. Assoc. Petroleum Geol., B., vol. 45, no. 5,
pp. 645-665. Defines and differentiates rock-stratigraphic, soil-strati-
graphic, bio-stratigraphic, time-stratigraphic, geologic-time, and geologic-
climate units.

AMSTUTZ, GERHARDT CHRISTIAN, 1960, Some basic concepts and thoughts on
the space- time analysis of rocks and mineral deposits in erogenic belts:
Geol. Rundschau, vol. 50, pp. 165-189. "The discovery, proof, and
acceptance of evolution of life was only one victory of the principle of
endogenesis over 'higher superstition 3 which clung ... to exogenesis. The
gradual departure of science from alchemistic concepts is essentially a
development from exo- toward endo-genesis, and of epi- toward syn-
genesis ..."

ANDERSON, CHARLES ALFRED, 1951, Older Precambrian structure in Arizona:
Geol. Soc. Am., B., vol. 62, no. 11, pp. 1331-1346. An explicit application
of the principle of simplicity to tectonic history is found on p. 1346. "Until
more precise correlations of the older Precambrian rocks in Arizona can
be made . . . the simplest explanation is that only one period of orogeny
. . . has occurred in Arizona during early Precambrian time."

ANDREE, KARL ERICH, 1938, Rezente und fossile Sedimente; Erdgeschichte mit
oder ohne Aktualitatslehre?: Geol. Rundschau, vol. 29, no. 3-5, pp. 147-
167. "Die Erdgeschichte ... die auch fur den Aktualitatsanhanger . . .
den Kern geologischer Forschung bildet, vermag der Aktualitatslehre
nicht zu entraten, ohne iiberhaupt den Boden unter den Fussen zu ver-
lieren, auf dem sie aufbaut."

ANDREWS, ERNEST CLAYTON, 1936, Some major problems in structural geology:
Linnean Soc. New South Wales, Pr., vol. 63, pts. 1-2, no. 275-276, pp.
iv-xl. "The student of the Newton and Darwin type seeks a simple ex-
planation of natural phenomena; he knows that there are many ways in
which, conceivably, a form such as a mountain range may have been
caused, whereas, in reality, it was produced in only one way . . ."

265

266 CLAUDE C. ALBRITTOX, JR.

APRODOV, V. A. 5 1961, Osnovnvje cherty filosofskago materializma v geologi-
cheskikh rabotakh M. V. Lomonosova (Basic features of philosophical
materialism in the geologic \vorks of M. V. Lomonosovj: Sovet. Geol.,
no. 12, pp. 3-13. Lomonosov's formulation of a materialistic method of
scientific investigation was as follows: through observations, form a theory;
through theory, check observations.

ARKELL, WILLIAM JOSCELYX, 1956, Species and species: Systematics Assoc.,
London, Pub. no. 2 5 pp. 97-99. As conceived by paleontologists, species
and genera are "purely artificial and subjective categories . . . Thus the
only logical criterion for the size and definition of the taxon in paleontology-
is its usefulness."

2 1956, Comments on stratigraphic procedure and terminology: Am. J. Sci.,

vol. 254, no. 8, pp. 457-46 7 . "In stratigraphic procedure, it is not what
terms an author uses that matters, but whether he knows what he is talking
about. If he does, he will be unambiguous however few and simple the
terms he employs, but if he does not, the more technical terms he uses,
and the more technical those terms are, the greater will be the danger of
confusion and misapprehension. 55

BACKLUND, HELGE G., 1941, Zum Aktualitatsprinzip: Geol. Rundschau, vol.
32, no. 3, pp. 394-397.

BAKER, HOWARD BIGELOW, 1938, Inductive logic and Lyellian uniformitarian-
ism: Michigan Acad. Sci., Sect. Geol. and Miner., pp. 1-5. "In the
inductive process the more hypotheses the better . . . Contrary to this
essential . . . the doctrine of uniformitarianism leads to poverty where
riches are desired.' 5

BARGHOORN, ELSO STERRENBERG, JR., 1953, Evidence of climatic change in the
geologic record of plant life, Chapter 20, pp. 235-248 in Harlow Shapley,
ed., Climatic change; evidence, causes, and effects: Cambridge, Harvard
Univ. Press, ix and 318 pp. "There are two basic assumptions for the
paleobotanical interpretation of climatic history. The first of these is that
plant groups of the past had environmental requirements similar to those
which they possess today . . . The second ... is that an environmental
complex of definable climatic and other physical conditions supports a
biotic population which is in general equilibrium with these conditions . . ."

BARKER, STEPHEN F., 1961, The role of simplicity in explanation, pp. 265-274
in Herbert Feigl and Grover Maxwell, eds., Current issues in the phil-
osophy of science: New York, Holt, Rinehart, and Winston, 484 pp. The
logical-empiricist view of explanation, "exaggerates the correlation be-
tween prediction and explanation, . . . exaggerates the contrast between
theoretical and observational terms in science, and ... by neglecting the
factor of simplicity it gives no real account of what it is for one explanation
to be a better explanation than another. 55

BIBLIOGRAPHY 267

2 1961, On simplicity in empirical hypotheses: Philosophy of Science, vol.

28, no. 2, pp. 162-171. ". . . the hypothesis that . . . fossils really are
remains of primeval organisms is a better hypothesis than is the hypothesis
that they are not and were created only recently. The former hypothesis
is better than the latter, not because it fits more facts or enables more pre-
dictions to be elicited; it is better because it is a simpler hypothesis."

BARRELL, JOSEPH, 1917, Rhythms and the measurement of geologic time:
Geol. Soc. Am., B., vol. 28, pp. 745-904. "The doctrine of uniformitarian-
ism has ignored the presence of age-long rhythms, and where they were
obtrusive has sought to smooth them out; but in so doing it has minimized
the differences between the present and the past, and the constant varia-
tions within that past. This doctrine should be looked on only as supplying
a beginning for investigation."

BARTH, THOMAS FREDRIK WEIBY, 1952, Theoretical petrology: New York, John
Wiley; London, Chapman and Hall, Ltd., viii and 387 pp. "Now that
experimental methods have led to the synthesis of minerals and rocks and
the determination of their thermodynamical constants, petrology has be-
come physico-chemistry applied to the crust of the earth."

BAULIG, HENRI, 1938, Questions de terminologie. I. Consequent, subsequent,
obs6quent; ou cataclinal, monoclinal, anaclinal?: J. Geomorphology, vol. 1 3
no. 3, pp. 224-229. "... one should never lose sight of the fact that theo-
retical schemes imply hypotheses, and that the appropriate genetic termin-
ology is applicable only to the extent that these hypotheses are demon-
strated to be correct, or else are explicitly admitted as a basis of discussion.
Otherwise it is better to employ purely descriptive terms."

2 1949, Causalit6 et finalit6 en g6omorphologie: Geog. Ann., Stockholm,

vol. 31, no. 1-4, pp. 321-324. Geomorphic explanations are commonly
colored by anthropomorphism a tacit ascription of motives and aspira-
tions to the mechanical agencies of erosion and deposition.

3 1950, William Morris Davis: master of method: Assoc. Am. Geographers,

Annals, vol. 40, no. 3, pp. 188-195. "Davis* s general method, as, in fact,
all geological inferences from the present to the past, suffers from a sort of
congenital weakness: for one term of the analogist argument, i.e., the
present, can not be taken to represent the 6 normal 5 condition of the
Earth."

BECKER, CARL LOTUS, 1932, The heavenly city of the eighteenth-century phil-
osophers: New Haven, Yale Univ. Press, Yale Paperbound Y-5, 168 pp.
"Much of what is called science is properly history, the history of biological
or physical phenomena. The geologist gives us the history of the earth;
the botanist relates the Me history of plants . . . We cannot properly know

268 CLAUDE C. ALBRITTOX, JR.

things as they are unless we know 6 how they came to be what they are' . . .
Historical mindedness is so much a preconception of modern thought that
we can identify a particular thing only by pointing to the various things it
successively was before it became that thing which it will presently cease
to be."

BECKNER, MORTON, 1959, The biological way of thought: New York, Columbia
Univ. Press, viii and 200 pp. "Scientists are most likely to have recourse
to historical explanation . . . when present phenomena seem explicable in
terms of a temporal sequence of past events.' 9

BELL, WILLIAM CHARLES, 1959, Uniformitarianism or uniformity: Am. Assoc.
Petroleum Geol., B., vol. 43, no. 12, pp. 2862-2865. "Paradoxically,
although we give lip-service to an awareness of isochronous but different
facies when we write theoretically, in practice most of our arguments are
peripheral to an almost axiomatic declaration that similarity implies
isochroneity and the greater the similarity the more probable the identity
in age."

BELOUSOV, V. V., 1938, "Teoriya zemli' 5 Dzhemsa Gettona (James Hutton's
''Theory of the Earth" J; Istoriya i Filosofiya Estest\ oznaniya, no. 7-8,
pp. 156-162. Hutton stated two general propositions which are important
to the philosophy of geology: (1) in reconstructing the past we can see and
foresee the results of changes but neither the beginning nor the end of the
processes which caused them, and (2) geologic time is vastly long in terms
of ordinary human perspectives.

BEMMELEN, REIN OUT WILLEM VAN, 1959, Die Methode in der Geologic: Geol.
Ges. \Vien, Mitt., vol. 53, pp. 35-52. "Unser irdisches System geochem-
ischer Prozesse ist nicht eine endlose Yerflechtung wiederkehrender Kreis-
laufe ... Es ist ein Teil einer kosmichen Evolution mit gerichtetem
Ablauf..."

2 1961, The scientific character of geology: J. Geol., vol. 69, no. 4, pp. 453-

463. Geology, essentially an historical science, differs from physics, chem-
istry, and biology- in that the possibilities for experiment are limited. The
principle of uniformity and the method of comparative ontology- are
examples of rules followed in geological practice, but the premises inherent
in these are not so firmly grounded as are the natural laws of physics and
chemistry. Thus the geologist must resort to the method of multiple work-
ing hypotheses to test the greatest number of presuppositions.

BERINGER, CARL CHRISTOPH, 1954, Geschichte der Geologic und des geolo-
gischen Weltbildes: Stuttgart, Ferdinand Enke, 158 pp.

BERKEY, CHARLES PETER, 1933, Recent development of geology as an applied
science: Am. Phil. Soc., Pr., vol. 72, no. 1, pp. 25-37.

BIBLIOGRAPHY 269

BERRY, EDWARD WILBER, 1925, On correlation: Geol. Soc. Am., B., vol. 36,
no. 1, pp. 263-277. "After all that can be said for them, it remains true
that classifications are conveniences for reference and mediums of ex-
change, and chronologic boundaries are quite as artificial as most political
boundaries and far more subjective."

2 1929, Shall we return to cataclysmal geology?: Am, J. Sci., 5th ser , vol.

17, pp. 1-12. "So much nonsense has been written on various so-called
ultimate criteria for correlation that many have the faith or the wish to
believe that the interior soul of our earth governs its surface history with a
periodicity like that of the clock of doom, and when the fated hour strikes
strata are folded and raised into mountains, epicontinental seas retreat,
the continents slide about, the denizens of the land and sea become dead
and buried, and a new era is inaugurated."

BETHUNE, Pierre de, 1953, Le cycle d 5 erosion: Rev. Questions Sci., an. 66,
vol. 124 (s. 5, vol. 14) f. 3, pp. 321-346. ". . . la tteorie si fcconde des
cycles d'6rosion ne constitue qu'une premiere approximation, et . . . seules
de patientes recherches comparatives permettront d'6difier la theorie
g6n6rale . . ."

2 1957, Un demi-sicle de g6omorphologie davisienne: Rev. Questions Sci.,

an. 69, vol. 128 (s. 5, vol. 18) f. 1, pp. 100-111. "La faiblesse cong&iitale
du davisianisme est dans cette m6thode a priori, e deductive,' de son raison-
nement. Nonobstant tout ce que les davisiens ont pu 6crire a ce sujet, la
vraie m6thode de la gomorphologie ... est Pinduction"

BEURLEN, KARL, 1935, Der Aktualismus in der Geologic, eine Klarstellung-
Zbl. Miner., 1935, Abt. B., no. 12, pp. 520-525. "Der starkste und nach
haltigste Verstoss gegen den Dogmatismus in der Geologic war die
Einfuhrung der aktualistischen Methode. . . . Wenn diese Methode dann
allerdings zu einem Forschungsprinzip gemacht wurde ... so war das aller-
dings eine unzulassige Verallgemeinerung und Umwertung der Methode
zu einem Dogma . . . Dadurch machte sich der Aktualismus selbst dessen
schuldig, was er zunachst bekampfte."

2 1935 1936, Zur Kritik des Aktualismus; I. Bedeutung und Aufgabe

geologischer Forschung; II. Das Klima des Diluviums; III. (with S.
Thiele) Das Experiment in der Tektonik und seine Bedeutung in der
Geologic: Zs. gesamte Naturw., vol. 1, pp. 23-36, 209-220; vol. 2, pp.
49-61.

BILLINGS, MARLAND PRATT, 1950, Stratigraphy and the study of metamorphic
rocks: Geol. Soc. Am., B., vol. 61, no. 5, pp. 435-448. In attacking field
problems, including those involving many metamorphic terranes, strati-
graphy and structure must be solved simultaneously.

BLACKWELDER, ELIOT, 1909, The valuation of unconformities: J. Geol., vol.
17, pp. 289-299. "The entire geological record ... is not to be conceived

2"0 CLAUDE C ALBRITTOX, JR.

of as a pile of strata, but as a dovetailed column of wedges, the uncon-
formities and rock systems being combined in varying proportions. The
former predominate in some places and periods, while the latter prevail in
others."'

BLAKE. RALPH M., DUCASSE, CURT J.. and MADDEN, EDWARD H., 1960,
Theories of scientific method; the Renaissance through the nineteenth
century: Seattle, University of Washington Press, iv and 346 pp. Critiques
of the principle of uniformity of nature as treated in the writings of Newton,
Hume, Herschel, Mill, and Jevons.

BLANC, ALBERTO CARLO, 1951, Cosmolyse et epistemologie non-cart6sienne, in
Sciences de la terre, XXI Cong, internal. Phil. Sci., Paris, 1949: Paris,
Hermann, pp. 105-122. Distinctive floras and faunas (including the spe-
cialized human races of so-called 'primitive cultures 5 ;, as the simplified
derivatives of originally more complex entities, are analogous to the hydro-
gen atom, which, according to the vie\\s of modern physics, is a simplified
end-product in the evolution of cosmic matter.

BLISS, HENRY EVELYN, 1952, A bibliographic classification: New York, H. W.
Wilson, 2 vols. Astronomy, geology, geography, and meteorology are
special natural sciences combining concrete and descriptive subject matters
with theoretical components derived from and dependent upon both
physics and chemistry.

BONDI, HERMANN, 1959, Science and structure of the universe: Manchester Lit.
and Phil. Soc., Mem. and Proc., vol. 101 (1958-1959), pp. 58-71. "I have a
suspicion (although I would not put it higher than that) that the statement
of uniformity makes sense in just such a universe as I think we have got,
which, though it is infinite in extent, yet is effectively finite owing to the
recessional velocities . . . The assumption [of uniformity] is by no means
clear in its meaning but empirically it seems by far the most fertile we can
make."

2 1961, Cosmology*: Cambridge Monographs on Physics, Cambridge Univ.

Press, 182 pp. "... in all our physics we have presupposed a certain uni-
formity of space and time; we have assumed that we live in a world that
is homogeneous at least as far as the laws of nature are concerned. Hence
the underlying axiom of our physics makes certain demands on the struc-
ture of the universe: it requires a cosmological uniformity."

BRADLEY, WILMOT HYDE, 1948, Limnology and the Eocene lakes of the Rocky
Mountain region: Geol. Soc. Am., B., vol. 59, no. 7, pp. 635-648. "But
what is the central theme peculiar to the science of geology that core
which is not derived from any of the sister disciplines? Perhaps you will
agree that it is the history and constitution of the earth."

BIBLIOGRAPHY 271

BRETSKY, PETER WILLIAM, JR., 1962, Barker on simplicity in historical geology:
Dallas, Southern Methodist Univ. Press, J. Graduate Research Center,
vol. 30, no. 1, pp. 45-47. "If Lyellian uniformitarianism ... be a kind of
notational simplicity, does it not then become trivial?"

BRIDGMAN, PERCY WILLIAMS, 1959, The way things are: Cambridge, Harvard
Univ. Press, x and 333 pp. Occam's Razor "appeals to me as a cardinal
intellectual principle ... I do not know what logical justification can be
offered for the principle. To me it seems to satisfy a deep seated instinct
for intellectual good workmanship. Perhaps one of the most compelling
reasons for adopting it is that thereby one has given as few hostages to the
future as possible and retained the maximum flexibility for dealing with
unanticipated facts or ideas."

BROGGI, JORGE ALBERTO, 1935, Los dominios de la geologia: Bol. Minas, Ind.
y Constr. (Lima, Peru) s. 3, vol. 7, pp. 13-18. "El tiempo actual en his-
toria, es un elemento de su constitucion futura; en geologia es la base de
las investigaciones sobre el pasado."

BROOM, ROBERT, 1932, Evolution as the palaeontologist sees it: South African
J. Sci., vol. 29, pp. 54-71. "As evolution has practically finished and can-
not be repeated unless all higher life is wiped off from the earth and a new
start made from the very beginning, we may perhaps conclude that man
is the end to which some power has guided evolution."

BROUWER, AART, 1957, On the principles of Pleistocene chronology: Geol. en
Mijnbouw (n.s.) vol. 19, pp. 62-68. "What is it that has induced geologists
to adopt paleontology as the base of geochronology? Certainly not its
reliability."

BROWN, BAHNGRELL WALTER, 1959, Preliminary study of stochastic terms used
in geology: Geol. Soc. Am., B., vol. 70, no. 5, pp. 651-654. Precision
would be gained in geological writing if definite ranges of probability
values were assigned to the following words expressing likelihood: impos-
sibly, improbably, equally likely, probably, undoubtedly, and possibly.

2 1961, Stochastic variables of geologic search and decision: Geol. Soc. Am.,

B., vol. 72, no. 11, pp. 1675-1685. Demonstrates that certain problems of
geologic search and decision can be treated stochastically.

BRYAN, KIRK, 1950, The place of geomorphology in the geographic sciences:
Assoc. Am. Geographers Annals, vol. 40, no. 3, pp. 196-208. "The first
and most natural application of geomorphic study is to the history of the
earth."

BUBNOFF, SERGE VON, 1937, Die historische Betrachtungsweise in der Geologic:
Geistige Arbeit, Berlin, vol. 4, no. 13, pp. 1-3. "Aktualismus bedeutet

272 CLAUDE C. ALBRITTON, JR.

nicht, dass alles immer ebenso war wie heute, sondern dass Reaktionen
(physikalische und biologische) derselben Grossenordnung von Raum und
Zeit unter gleichen Voraussetzungen gleich verlaufen."

2 (Hans Cloos and Georg Wagner), 1943, \Varum Geologic?: Beitr. Geol.

Thiiringen, vol. 7, no. 4-5, pp. 191-204. "Diese Verknupfung naturwis-
senschaftlicher und historischer Denkungs- und Untersuchungsart und
diese Lebensnahe und 'Bodenstandigkeit' lassen die Geologic fur die
Schule daher besonders wichtig und not\vendig erscheinen . . ."

3 1954, Grundprobleme der Geologic, 3rd ed.: Berlin, Akademie-Verlag,

\ii and 234 pp. "Die Grundfragen, welche dazu formuliert werden miissen,
waren etwa: Was ist unsere \Vissenschaft, welche Wege verfolgt sie, welche
Fehler sind in ihrem methodischen Aufbau moglich, was sind ihre End-
ziele?"

BUCHER, \\ T ALTER HERMANN, 1936, The concept of natural law in geology:
Ohio J. Sci., vol. 36, no. 4, pp. 183-194. "Geology is peculiarly dual in
its aims: on the one hand it is concerned with what happened once at a
certain place, in individual mines, mountains, regions. Interest that centers
on individuals is history, not science. As a science, geology is concerned with
the typical that finds expressions in generalizations, whether they be called
laws or something else."

2 1941, The nature of geological inquiry and the training required for it:

Am. Inst. Min. Met. Eng., Tech. Pub. 1377, 6 pp. "The typical 'geologi-
cal' processes cannot be studied directly by laboratory methods but only
indirectly by their results; that is, by the methods of the historical sciences.
. . . the greater part of all geological work is not primarily concerned with
'timeless 5 knowledge but with concrete, 'time-bound' reality. It deals not
with ore bodies, but with this ore body; not with valleys in general, but
with that valley."

3 1957, The deformation of the earth's crust; an inductive approach to the

problems of diastrophism: New York, Hafner, xii and 518 pp. Forty-six
laws relating to diastrophism are formulated.

BULOW, KURD VON, 1943, Geschichte und Zukunft der Formationstabelle:
Neues Jb. Miner., GeoL, u. Palaont., 1943, Abt B., no. 5, pp. 116-130.
"So entstand die Tabelle nicht eigentlich als Gliederung, Einteilung eines
Ganzen, sondern dieses Ganze wurde erst durch die stratigraphische Klein-
arbeit gefugt."

2 1960, Der Weg des Aktualismus in England, Frankreich und Deutschland:

GeoL Ges. Deut. Dem. Rep., Ber., vol. 5, no. 3, pp. 160-174. Actualism
has been interpreted in two different ways: as a presupposition concerning
the steady course of nature, and as a method for applying the experience
of the present to the reconstruction of the past. The present tendency
among geologists is to accept actualism as a useful method rather than as
an article of faith.

BIBLIOGRAPHY 273

BUNGE, MARIO AUGUSTO, 1961, The weight of simplicity in the construction
and assaying of scientific theories: Philosophy of Science, vol. 28, no. 2,
pp. 120-149. "Historical theories such as those of geology, evolution
and human society have a high explanatory power but a small predictive
power, even counting retrodictions."

CAILLEUX, ANDRE, and TRICART, JEAN, 1961, Id6alisme, mat&ialisme et ac-
celeration: Rev. Geomorph. Dynamique, vol. 12, no. 1, pp. 1-2. A defense
of materialism, in the sense of an empirical approach to geological prob-
lems.

CAMP, CHARLES LEWIS, 1952, Geological boundaries in relation to faunal
changes and diastrophism: J. Paleont, vol. 26, no. 3, pp. 353-358. "Noth-
ing much can happen to our facts, but fitting these facts together to erect
. . . systems of biological classification and geological age-dating is an un-
ending struggle. Awaiting possible refinements in absolute age-dating, we
are often constrained to fit the facts together, and let the devil take the
age boundaries of the geological time scale the facts and the boundaries
don't always coincide."

CANNON, WALTER F., 1960, The uniformitarian-catastrophist debate: Isis,
vol. 51, pt. 1, no. 163, pp. 38-55. ". . . if Darwin was deeply indebted to
Charles Lyell for the method of accounting for large changes by summing
up small changes over immense periods of time, nevertheless he did not
accept the general Uniformitarian account of the history of nature. Evolu-
tion by means of natural selection involves the acceptance of the idea that
some sort of cumulative development is demonstrated by geological and
biological evidence and it is just this idea that Uniformitarianism con-
sistently denied."

2 1960, The problem of miracles in the 1830's: Victorian Studies, vol. 4,

no. 1, pp. 5-32. "With their stubborn insistence on the historical nature
of the cosmos . . . the Catastrophists forced speculation away from Lyell's
unprogressive position and kept a developmental view of the world alive,
so that today we popularly describe the history of the world as Whewell
saw it (but without his miracles), not as Lyell saw it."

3 1961, John Herschel and the idea of science: J. History of Ideas, vol. 22,

no. 2, pp. 21 5-239. "It might seem to a modern man that geology is neces-
sarily a 'historical' science, but Uniformitarianism attempted to make it
as 'unhistorical, 5 that is to say, as much without development, as possible.
LyelPs Principles of Geology was the last great codification of a non-develop-
mental cosmography."

4 1961, The impact of Uniformitarianism; two letters from John Herschel to

Charles Lyell, 1836-1837: Am. Phil. Soc., Pr., vol. 105, no. 3, pp. 301-314.
HerscheFs belief that the Uniformitarian-Catastrophist debate was at base
inconsequential was incorrect, "since the Uniformitarian antidevelop-

2"4 CLAUDE C. ALBRITTOX, JR.

mental view of the world was quite antithetic to the Catastrophist insistence
on a world developing geologically from a primitive chaos/ 1

CARXAP, RUDOLF, 1955, Logical foundations of the unity of science: Chicago,
Univ. Chicago Press, Int. Encyc. Unified Sci., vol. 1, pts. 1-5, pp. 42-62.
"Let us take 'physics' as a common name for the non-biological field of
science, comprehending both systematic and historical investigations with-
in this field, thus including chemistry, mineralogy, astronomy, geology
(which is historical), meteorology, etc."

CASSIRER, ERNST, 1953, An essay on man; an introduction to a philosophy of
human culture: Garden City, N. Y., Doubleday, 294 pp. The same logical
structure characterizes the historical thought of the historian, geologist,
and paleontologist, but the historian has the unique task of interpreting
the symbolic content of human documents and monuments. (See Chapter
10, Historj.)

CAYEUX, LUCIEN, 1941, Causes anciennes et causes actuelles en geologic: Paris,
Masson, 79 pp. "II est necessaire de reserver une place des Causes
anciennes, & c&te des Causes actuelles, dans l'6tude des formations s6di-
mentaires de P6corce terrestre, si Ton veut faire appel a toutes les lumiferes,
susceptibles de nous en donner 1' intelligence."

CHAMBERLIN, THOMAS CHROWDER, 1897, The method of multiple working
hypotheses: J. Geol., vol. 5, pp. 837-848. c ln developing the multiple
hypotheses, the effort is to bring up into view every rational explanation
of the phenomenon in hand and to develop every tenable hypothesis rela-
tive to its nature, cause or origin, and to give to all of these as impartially
as possible a working form and a due place in the investigation."

2 1898, The ulterior basis of time divisions and the classification of geologic

history: J. Geol., vol. 6, pp. 449-462. "The most vital problem before the
general geologist today is the question whether the earth's history is
naturally divided into periodic phases of world-wide prevalence, or
whether it is but an aggregation of local events dependent upon local
conditions uncontrolled by overmastering agencies of universal domi-
nance."

3 1904, The methods of the earth sciences: Pop. Sci. Monthly, vol. 66, pp.

66-75. "In some sense the earth sciences must come to comprehend the
essentials of all the sciences. At least as much as any other scientists we are
interested in the fundamental assumptions of all the sciences and in their
consistent application."

4 1907, Editorial: J. Geol., vol. 15, pp. 817-819. "The cheapest device for

making the largest show of quasi-results in technical garb with the least
investment of scientific capital known to our profession is found in giving
new names to known formations."

BIBLIOGRAPHY 275

CHORLEY, RICHARD J., 1957, Illustrating the laws of morphometry: Geol.
Mag., vol. 94, no. 2, pp. 140-150. "The laws of morphometry are here
illustrated from three regions of maturely dissected sandstone terrain,
lacking in differential gross structural control, and it is discovered that
uniform, dimensionless ratios hold for their geometry."

CLARKE, JOHN MASON, 1917, The philosophy of geology and the order of the
state: Geol. Soc. Am., B., vol. 28, pp. 235-248. "Nature makes for the
individual . . . The ants are nature's . . . highest performance in commu-
nistic effort and in cooperative achievement."

CLEMENTS, FREDERIC EDWARD, and SHELFORD, VICTOR E., 1939, Bio-ecology:
New York, John Wiley, vii and 425 pp. "... the key to the past is fash-
ioned by the present, to use these terms in their everyday significance. On
the other hand, the present is the sole heir to the past, and no adequate
understanding of it is possible without tracing the continuity of develop-
mental processes from the one to the other."

CLEUGH, MARY FRANCES, 1937, Time and its importance in modern thought:
London, Methuen, viii and 308 pp. "'The Past' is a curious entity. Al-
though changeless ... it yet has a peculiar habit of growing. The possible,
the general, the abstract: all these are subject to logical determinations
but . . . when we pass from possibility to actuality, when we come to events
happening in time, we have left that ideal realm of logic, and have intro-
duced a radical contingency into the universe."

CLOUD, PRESTON ERCELLE, JR., 1959, Paleoecology retrospect and prospect:
J. Paleont., vol. 33, no. 5, pp. 926-962. A review of fundamental concepts
and problems, with an extensive bibliography.

COLBERT, EDWIN HARRIS, 1953, The record of climatic changes as revealed by
vertebrate paleoecology, Chapter 21, pp. 249-271, in Harlow Shapley,
ed., Climatic change; evidence, causes, and effects: Cambridge, Harvard
Univ. Press, ix and 318 pp. "We assume that ecological conditions (and
by extension, climatic conditions as well) were thus and so because of
the presence of certain animals in the sediments. This assumption rests
upon a basic precept that must be accepted at the outset, if our interpre-
tation of past climates upon the evidence of vertebrate paleoecology is to
have order and validity. The precept is that within wide limits the past
is to be interpreted in terms of the present."

COLLINGWOOD, ROBIN GEORGE, 1956, The idea of history: New York, Oxford
Univ. Press, xxvi and 339 pp. "... whereas science lives in a world of
abstract universals, which are in one sense everywhere and in another
nowhere, in one sense at all times and in another at no time, the things

2"6 CLAUDE C. ALBRITTOX, JR.

about which the historian reasons are not abstract but concrete, not uni-
versal but individual, not indifferent to space and time but having a where
and a when of their own, though the where need not be here and the when
cannot be now."

2 1960, The idea of nature. New York, Oxford Univ. Press, viii and 183 pp.

"I conclude that natural science as a form of thought exists and always
has existed in a context of history, and depends on historical thought for
its existence. From this I venture to infer that no one can understand
natural science unless he understands history: and that no one can answer
the question what nature is unless he knows what history is. This is a
question which Alexander and Whitehead have not asked. And that is
why I answer the question, c \Vhere do we go from here?' by saying, 'We
go from the idea of nature to the idea of history/ *'

COOMBS, D. S., 1957, The growth of the geological sciences^ an inaugural lec-
ture delivered before the University of Otago on 5 July, 1956: Dunedin,
New Zealand, Univ. Otago. 19 pp.

COOPER, GUSTAV ARTHUR, 1958, The science of paleontology: J. Paleont, vol.
32, no. 5, pp. 1010-1018. Paleontology is essentially a biological science
embracing six disciplines: morphologv, taxonomy, evolution, distribution
of fossils in space, paleoecology, and correlation.

COTTA, BERNHARD vox, 1874, Die Geologic der Gegenwart: Leipzig, J. J.
Weber, xii and 450 pp. "Die Mannigfaltigkeit der Erscheinungsformen ist
eine nothwendige Folge der Summierung von Resultaten aller Einzel-
vorgange, die nacheinander eingetreten sind . . ."

CRICKMAV, COLIN HAYTER, 1959, A preliminary inquiry into the formulation
and applicability of the geological principle of uniformity: Calgary,
Alberta, privately printed, iii and 50 pp. "Uniformity as a principle was
originally conceived as a rupture with the older notion of geomorphic
immutability, and was therefore in essence a theory of uniformity of
change . . . The law of uniformity of the essence of things geologic must
include, in order to be general, the possibility of wider variation of causes
and effects than any so far assigned."

DALY, REGINALD ALD WORTH, 1910, Some chemical conditions in the pre-
Cambrian ocean: llth Internat. Geol. Cong., Stockholm, C. R., vol. 1,
pp. 503-509. "... a strictly uniformitarian view of the ocean's history
must be in error; ... its evolution has, on the whole, meant progress
from a relatively fresh condition to the present saline condition . . ."

2 1945, Biographical memoir of William Morris Davis: Nat. Acad. Sci.

Biog. Mem., vol. 23, llth Mem., pp. 263-303. "Valuable as it is, the
scheme of the erosion cycle is not so important for research in earth science

BIBLIOGRAPHY 277

as the underlying philosophy, which makes deduction no whit inferior to
induction in the tool-chest of the naturalist."

DAVIS, WILLIAM MORRIS, 1895, Bearing of physiography on uniformitarianism
(Abs.): Geol. Soc. Am., B., vol. 7, pp. 8-11. "Uniformitarianism, reason-
ably understood, is not a rigid limitation of past processes to the rates of
present processes, but a rational association of observed effects with compe-
tent causes. Events may have progressed both faster and slower in the
past than during the brief interval which we call the present, but the past
and present events differ in degree and not in kind."

2 1904, The relations of the earth sciences in view of their progress in the

nineteenth century: J. Geol., vol. 12, pp. 669-687. "Geology objectively
considered is not merely one of the earth's sciences; it is the whole of them:
it is the universal history of the earth."

3 1926, The value of outrageous geological hypotheses: Science, n.s., vol. 63,

pp. 463-468. "The very foundation of our science is only an inference;
for the whole of it rests on the unprovable assumption that, all through
the inferred lapse of time which the inferred performance of inferred geo-
logical processes involves, they have been going on in a manner consistent
with the laws of nature as we know them now."

4 1926, Biographical memoir Grove Karl Gilbert, 1843-1918: Nat. Acad.

Sci. Biog. Mem., vol. 21, 5th Mem., v and 303 pp. Gilbert's laws of erosion
are discussed on pp. 49-50, and his views of scientific method on pp. 145-
148.

5 1954, Geographical essays (edited by Douglas Wilson Johnson): New

York, Dover Publications, vi and 777 pp.

DAWSON, Sir JOHN WILLIAM, 1894, Some recent discussions in geology: Geol.
Soc. Am., B., vol. 5, pp. 101-116. "Dead materialistic uniforrnitarianism,
should it ever become the universal doctrine of science, would provoke a
reaction in the human mind which would be itself a cataclysm."

DEER, WILLIAM ALEXANDER, 1953, Trends in petrology: Manchester Lit. and
Phil. Soc., Mem. and Pr., vol. 94 (1952-1953) pp. 63-92. The petrologist
"is presented with end products that have attained their present state
after many changes both in space and time. The basic problem is to un-
ravel this complex and ever-changing time-space pattern. Although this
is the essence of all science, it is much more the everyday province of
geology than of the other sciences. Most sciences start with the raw ma-
terial; the geologist is presented with the finished product and, unlike the
biologist, his material is not alive but dead. It is his job to make it live."

DEEVEY, EDWARD SMITH, JR., 1944, Pollen analysis and Mexican archeology:
an attempt to apply the method: Am. Antiquity, vol. 10, no. 2, pp. 134-
149. "On the whole, then, Occam's principle of economy of hypotheses

2^8 CLAUDE C. ALBRITTOX, JR.

compels us to hold in abeyance, for the moment, so detailed an applica-
tion of the 'climatic factor" in human history, but this would not destroy
the usefulness of slight changes of vegetation or climate as chronological
markers in a pollen sequence."

2 1953, Paleolimnology and climate, Chapter 22, pp. 273-318 in Harlow

Shapley, ed., Climatic change; evidence, causes, and effects: Cambridge,
Harvard Univ. Press., ix and 318 pp. fcS In the study of lakes, which are
the eco-systems that lend themselves most readilv to the kind of analysis
that does not lose sight of their essential wholeness, the geologist and the
ecologist can collaborate most happily. One knows that the present is
the key to the past, the other knows that the past is the key to the present,
and each can think of the other as the captive expert. 5 '

DE GOLYER, EVERETTE LEE, 1948, Science a method, not a field: Norman,
Univ. of Oklahoma Press, 15 pp. "... we may be somewhat pragmatical
and accept as truth that which is conformable with fact \\ithin our present
understanding.'"

DEM AY, ANDRE, 1951, Observation, interpretation et th6orie en geologic, in
Sciences de la terre, XXI Cong. Internat. Phil. Sci., Paris, 1949: Paris,
Hermann, pp. 41-54. In geology, where the more difficult problems are
those of genesis or evolution, a theory generally attempts to reconstruct a
sequence of phenomena and to disclose in these phenomena the workings
of mechanical, physical, or chemical laws.

DINGLE, HERBERT, 1955, Philosophical aspects of cosmology, in Arthur Beer,
ed., Vistas in astronomy: Oxford, Pergamon Press, vol. 1, pp. 162-166.
". . . we cannot grant the inaccessible past freedom to sow its wild oats as
it pleases . . .; it must to some extent conform to the pattern on which we
organize the present behavior of the universe."

DYLIK, JAN, 1953, Caracteres du developpement de la g6omorphologie mo-
derne: Soc. Sci. et Lettres L6dz, CL III, B., vol. 4, no. 3, 40 pp.

EBERT, HEINZ, 1953, Petrologie im kristallinen Praekambrium, genetisch oder
historisch?: 19th Internat. Geol. Cong., Algiers, 1952, C.R., sec. 1, fasc. 1,
pp. 81-87. "With respect to the e precambrian shields' certain petrological
mistakes remain hidden because of the widely spread misunderstanding
that the genetical viewpoint of petrologists is [the] equivalent of the
historical one of geologists. . . . Instead of attributing a genetical 'expla-
nation' to rocks, one must consider them as products of a long and com-
plicated evolution . . ."

EISLEY, LOREN COREY, 1958, Darwin's century; evolution and the men who
discovered it: Garden City, N. Y., Doubleday, xvii and 378 pp. "If there

BIBLIOGRAPHY 279

is one mind that deserves to rank between the great astronomical geniuses
of the seventeenth century and Charles Darwin in the nineteenth, it is
James Hutton ... He discovered an intangible thing against which the
human mind had long armored itself. He discovered, in other words,
time time boundless and without end, the time of the ancient Easterners
but in this case demonstrated by the very stones of the world ..."

EMMONS, EBENEZER, 1855, American geology, containing a statement of the
principles of the science with full illustrations of the characteristic American
fossils, with an atlas and a geological map of the United States, vol. 1,
pt. 1 : Albany, N. Y., Sprague, J. Munsell, xvi and 194 pp. Cw Each period
then is a fragment; but the present is a greater fragment than all the past
put together."

ENGELS, FRIEDRICH, 1940, Dialectics of nature; Translated and edited by
Clemens Dutt, with a preface and notes by J. B. S. Haldane: New York,
International Publishers, xvi and 383 pp. "The defect of Lyell's view
at least in its first form lay in conceiving the forces at work on the earth
as constant, both in quality and quantity. The cooling of the earth does
not exist for him; the earth does not develop in a definite direction but
merely changes in an inconsequent fortuitous manner."

FAIRCHILD, HERMAN LE ROY, 1904, Geology under the planetesimal hypothesis
of earth origin (with discussion): Geol. Soc. Am., B., vol. 15, pp. 243-266.
"Geologists have been too generous in allowing other people to make their
philosophy for them."

FEIGL, HERBERT, 1943, Logical empiricism, pp. 373-416 in Dagobert David
Runes, ed., Twentieth Century Philosophy: New York, Philosophical
Library, 571 pp. "If all a priori knowledge is analytic, then we cannot
deduce a synthetic assertion, like the principle of the uniformity of nature,
from a priori premises. And if we try to validate induction on the basis of
its certainly eminent success in the past, we are simply making an induction
about induction and thus presuppose the very principle we set out to
prove."

FEUER, LEWIS SAMUEL, 1957, The principle of simplicity: Philosophy of Science,
vol. 34, pp. 109-122. The metascientific and methodologic principles of
simplicity are independent concepts, the latter being no more than a
special case of the principle of verifiability.

2 1959, Rejoinder on the principle of simplicity: Philosophy of Science,

vol. 26, pp. 43-45. "... if we are going to make scientific statements
concerning the distant past or future, the only way we can do so is by an
extrapolative use of the known laws of nature. As applied to the past, this
is precisely what historical geology does."

280 CLAUDE C. ALBRITTOX, JR.

FISHER, Sir RONALD AYLMER, 1954, The expansion of statistics: Am. Scientist,
vol. 42, no. 2, pp. 2"5-282, 293. Some of the greatest advances of the
nineteenth century were accomplished by the application of statistical
ideas. A case in point is L veil's subdivision of the Tertiary sequence ac-
cording to the percentage of fossil species with living descendants. But
the statistical argument "by which this revolution in geological science
was effected was almost immediately forgotten ... It had served its pur-
pose; the ladder by which the height had been scaled could be kicked
down."

FLETT, Sir JOHN SMITH, 1940, Pioneers of British geology: Roy. Soc. New-
South Wales, J. Pr., 1939 5 vol. 73, pt. 2, pp. 41-66. ". . . though it is recog-
nised that there are limitations to its application, the general principle
holds good that the explanation of the past history of the earth must first
be sought in the processes that we see going on around us at the present
day/ 5

FRANKEL, CHARLES, 1957, Explanation and interpretation in history: Phil-
osophy of Science, vol. 24, pp. 137-155. ". . . it seems to be the case some-
times that when we ask for an explanation of a given phenomenon, what
we want, and are satisfied to get, is an account of the stages of a process,
the last stage of which is the phenomenon in the shape in which it exhibits
those traits about which we have asked our question. This is one of the
stable and accepted meanings of "explanation* in ordinary usage."

FREYBERG, BRUNO VON, et al., 1938, Fragen der geologischen Ausbildung:
Deut. Geol. Ges., Zs., vol. 90, no. 3, pp. 148-166. "Fur jede Ausbildung
muss leitend sein, dass es Erdgeschichte ist, was wir treiben, keine Erd-
beschreibung."

Fu, C. Y., 1948, Methods and problems of geophysics: Chinese Geophys. Soc.,
J., vol. 1, no. 1, pp. 1-15. "There are ... scientists who believe that
natural laws tend to be simple and natural phenomena . . . inherently
uniform. It is perhaps wise to aim at simplicity, but there is no a priori
reason why nature should be so characterized. The fact is that simplicity
and complexity alternate as science progresses . . ."

GALBRAITH, F. W., ed., 1961, What is geology?: Univ. Arizona, Studies in
Geology, vol. 1, 36 pp. Contains eleven articles on the scope of geology
and the interrelationship of the principal geological specialties.

GALLIE, WALTER BRYCE, 1955, Explanations in history and the genetic sciences:
Mind, vol. 64, pp. 160-180. "... a characteristically genetic explanation
emphasizes the one-way passage of time what came earlier explains, in
the genetic sense, what came later, and not vice versa. In other words the

BIBLIOGRAPHY 281

prior event is not taken, in conjunction with certain universal laws, to
constitute both a sufficient and a necessary condition of the occurrence of
the subsequent event."

GEIKIE, Sir ARCHIBALD, 1893, Geological change, and time: Smithsonian Inst.,
Ann. Kept., 1892, pp. 111-131. "Lord Kelvin is willing, I believe, to
grant us some twenty millions of years, but Professor Tait would have us
content with less than ten millions . . . After careful reflection on the sub-
ject, I affirm that the geological record furnishes a mass of evidence which
no arguments drawn from other departments of nature can explain away,
and which, it seems to me, can not be satisfactorily interpreted save with
an allowance of time much beyond the narrow limits which recent physical
speculation would concede."

2 1897, The founders of geology: London, Macmillan, x and 297 pp. Traces

the evolution of geological thought from 1750 to 1820.

GIGNOUX, MAURICE, 1951, Le rdle jou6 par les sciences de la terre dans nos
r6presentations de la mature, in Sciences de la terre, XXI Cong. Internal.
Phil. Sci., Paris, 1949: Paris, Hermann, pp. 123-129. The physical and
dynamical properties which determine the mechanics of the deformation
which a given substance will exhibit vary according to the scale of time
employed by the investigator.

GILBERT, GROVE KARL, 1886, The inculcation of scientific method by example,
with an illustration drawn from the Quaternary geology of Utah: Am. J.
Sci., 3d ser., vol. 31 (whole no. 131), pp. 284-299. A natural phenomenon
generally has multiple antecedents and consequents, the linear chains of
which are plexiform. To discover the antecedents of phenomena, hy-
potheses are invented and tested. The reasoning involved in devising and
testing hypotheses is analogical.

2 1896, The origin of hypotheses, illustrated by the discussion of a topo-
graphic problem: Science, n.s., vol. 3, pp. 1-13. ". . . hypotheses are
always suggested through analogy. The unexplained phenomenon on
which the student fixes his attention resembles in some of its features
another phenomenon of which the explanation is known."

GILLISPIE, CHARLES COULSTON, 1951, Genesis and geology, a study in the rela-
tions of scientific thought, natural theology and social opinion in Great
Britain, 1790-1850: Cambridge, Harvard Univ. Press, xiii and 315 pp.
"Geology . . . was the first science to be concerned with the history of
nature rather than its order . . . That its historical character made geology
a different sort of science was appreciated from the beginning of its de-
velopment. Doubts were sometimes expressed as to whether it could prop-
erly be called a science at all. Since the geologist, like the historian,
had to rely largely on ancient relics and monuments of change, his conclu-

282 CLAUDE C. ALBRITTOX, JR.

sions were thought to be debatable in a way that those of the physicist,
for example, were not."

GILLULY, JAMES, 1949, Distribution of mountain building in geologic time:
Geol. Soc. Am., B., vol. 60, no. 4, pp. 561-590. "... I can see no grounds
whatever for assuming any increased tempo for diastrophism during post-
Lipalian time. The uniformitarianism of Lyell seems not yet to require any
of the amendments that have been suggested."

GOGUEL, JEAN, 1948, La place de la g6ophysique parmi les disciplines geo-
logiques: Geol. Applique et Prosp. Min., no. 2, pp. 123-135. "Certes
dans 1'ere de la gtophysique nous faisons des mesures, mais ce que nous
mesurons, ce n'est pas, ce n'est jamais ce que nous voudrions deviner."

2 1951, La geologic, science naturelle ou science physique?, in Sciences de la

terre, XXI Cong. Internat. Phil. ScL, Paris, 1949: Paris, Hermann,
pp. 7-15. From the physical sciences, geology is distinguished by its his-
torical method. The historical reconstructions of geology must, however,
be in accord with the physical, chemical, and mechanical laws, which are
presumed to be permanent.

GOLD, THOMAS, 1956, Cosmology-, in Arthur Beer, ed.. Vistas in astronomy:
London and New York, Pergamon Press, vol. 2, pp. 1721-1726. "The
postulate of simplicity is sometimes so firmly in our minds that it is hard to
recognize; but it is quite certain that it is a postulate of the greatest im-
portance and that without it scientific progress, through the generalization
of known facts, would be impossible."

GOODCHILD, JOHN GEORGE, 1896, Some geological evidence regarding the age
of the earth: Roy. Phys. Soc. Edinburgh, Pr., vol. 13 (1896-1897), pp. 259-
308. ". . . the fundamental idea which a geologist steadily keeps in mind
is that all changes, physical and biological, which the records of the rocks
inform us have taken place on the Earth in the Past, can only be under-
stood and properly interpreted by reference to changes of the same nature
which are known to be in progress during the Present. This, of course,
does not imply absolute uniformitarianism (as this is commonly under-
stood) but allows for catastrophism in certain exceptional cases, along with
normal uniformity of action in the rest."

GOODMAN, NELSON, 1943, On the simplicity of ideas: J. Symbolic Logic, vol. 8,
no. 4, pp. 107-121. "The most economical idea, like the most economical
engine, is the one that accomplishes most by using least."

2 1955, Fact, fiction and forecast: Cambridge, Harvard Univ. Press., 126 pp.

"The typical writer begins by insisting that some way of justifying predic-
tions must be found; proceeds to argue that for this purpose we need some
resounding universal law of the Uniformity of Nature, and then inquires

BIBLIOGRAPHY 283

how this universal principle itself can be justified . . . Such an invention
. . . seldom satisfies anyone else; and the easier course of accepting an
unsubstantiated and even dubious assumption much more sweeping than
any actual predictions we make seems an odd and expensive way of
justifying them."

3 1958, The test of simplicity: Science, vol. 128, no. 3331, pp. 1064-1069.

"Nothing could be more mistaken than the traditional idea that we first
seek a true system and then, for the sake of elegance alone, seek a simple
one. We are inevitably concerned with simplicity as soon as we are con-
cerned with system at all; for system is achieved just to the extent that the
basic vocabulary and set of first principles used in dealing with the given
subject matter are simplified."

GORDEEV, D. I., 1960, Znachenije filosofskikh trudov V. I. Lenina dlya geologii
(The meaning of V. I. Lenin's philosophical works for geology): Moscow
Univ., Vest., ser. 4, GeoL, no. 4, pp. 3-7. Geology, more than any other
science, deals with cyclic phenomena; but this same periodicity, if not
approached from Lenin's viewpoint, will acquire a metaphysical and anti-
historical character, as can be seen in certain foreign writings.

2 1961, Stikhiinaya materialisticheskaya dialektika v geologicheskikh so-

chineniyakh M. V. Lomonosova (Dialectical materialism in the works of
M. V. Lomonosov): Moscow Univ., Vest., Geol., no. 5, pp. 7-26. In
1763, Lomonosov introduced the principle of actualism in geology. Be-
cause he perceived that the earth is evolving in a definite direction, his
uniformitarianism was more substantial than Lyell's.

GORDON, WILLIAM THOMAS, 1934, Plant life and the philosophy of geology:
British Assoc. Adv. Sci., Rept., 1934, pp. 49-82. "Taking everything we
know into consideration, the general consensus of opinion is that plants
do afford an index of climatic changes, and that these changes have been
very considerable in past time."

GOUDGE, THOMAS ANDERSON, 1958, Causal explanations in natural history:
British J. Phil. Sci., vol. 9, no. 35, pp. 194-202. "... the model of a hier-
archical deductive system, so often presented as the ideal to which the
theoretical part of every science should approximate, is not relevant to the
sciences concerned with natural history, however much it is relevant to
non-historical sciences."

2 1961, The genetic fallacy: Synthese, vol. 13, no. 1, pp. 41-48. Geology,

paleontology and other studies with a concern for historical development
regularly and justifiably make use of genetic explanations. Fallacies are
introduced in such explanations when temporal order is confused with
logical order, when it is assumed that a genetic explanation says all there
is to say about the phenomenon hi question, when it is assumed that the
phenomenon to be explained is simply the summation rather than the out-

284 CLAUDE C. ALBRITTOX, JR.

come of its antecedent stages, and when a trivial but necessary condition
of a highly developed state is treated as if it were a sufficient condition of
that state.

3 1961, The ascent of life: a philosophical study of the theory of evolution:

Toronto, Univ. Toronto Press, 263 pp. *\ . . the doctrine of evolution
would fail to be intelligible unless the uniformiiarian principle describes
what is the case.'" 1

GOULD, R. P , 195 7 , The place of historical statements hi biology: British J
Phil. Sci., vol. 8, pp. 192-210. "Retrodictions are obviously not as con-
vincing as predictions as confirmation or falsification of an explanatory
hypothesis because a Statement about the Past (which is largely hypo-
thetical) is being used as though it were an observation record."

GRABAU, WARREN EDWARD, 1960, Geology as an historical tool: Gulf Coast
Assoc Geol. Societies, Tr., vol. 10, pp. 87-91. As a reconstructive art,
geologv is potentially applicable to any field of human effort concerned
with the reconstruction of the past. Examples are given of historical events
in the Civil War which are explicated in terms of geological evidence.

GREENE, JOHN COLTON, 1959, The death of Adam; evolution and its impact
on Western thought: Ames, Iowa, Iowa State Univ. Press, 338 pp. "... the
explanation of terrestrial phenomena must be sought in the everyday
workings of nature, not in cataclysmic events. This was the basic principle
of umforrmtariamsm, the foundation stone of modern geology . . ."

GRENE, MARJORIE, 1958, Two evolutionary theories: British J. Phil. Sci.,
vol. 9, no. 34, pp. 110-127; no. 35, pp. 185-193. "... instead of Ockham's
razor we might adopt as a test of theories of evolution the opposite prin-
ciple: that entities, or more generally perhaps aspects of reality . . . should
not be subtracted beyond what is honest. In the light of this principle, we
should ask of any theory of evolution, does it pretend to do without con-
cepts which in fact it does not do without."

GRESSLY, AMANZ, 1838, Observations geologiques sur le Jura soleurois: Allgem.
Schweiz. Ges. f. d. gesammten Naturwiss., Neue Denkschr., vol. 2, 349
pp. "Et d'abord il est deux faits principaux, qui charact&isent partout
les ensembles de modifications que j'appelle facies ou aspects de terrain:
Fun consiste en ce que tel ou tel aspect petrographique fun terrain quelconque
suppose necessairement, partout oil il se rencontre, le meme ensemble paleontologique;
Fautre, en ce que tel ou tel ensemble paleontologique exclut rigoureusement des
genres et des especes de fossiles frequents dans d 9 autres facies"

GRIFFITHS, JOHN CEDRIG, 1960, Aspects of measurement in the geosciences:
Mineral Industries, vol. 29, no. 4, pp. 1, 4-5, 8. "The natural sciences lie

BIBLIOGRAPHY 285

somewhere between the 'precise' sciences and the 'imprecise' sciences and
possess some of the advantages and disadvantages of both; indeed, because
of this transitional position, the procedures utilized in the natural sciences
may benefit from an analysis using logical procedures from both 'hard'
and 'soft' sciences."

GUNTER, GORDON, 1953, The development of ecology and its relationship to
paleontology: Texas J. Sci., vol. 5, no. 2, pp. 137-147. ". . . all paleo-
ecology is based upon the fundamental assumption that any given group
of organisms lived in the past in an environment essentially similar to
that in which they or their counterparts live today. The validity of this
assumption is borne out by the fact that students of entirely different
groups of organisms check each other in their paleoecological interpreta-
tions."

GUTENBERG, BENO, 1937, Geophysics as a science: Geophysics, vol. 2, no. 3,
pp. 185-187.

HAARMANN, ERICH, 1935, Urn das geologische Weltbild, malleo et mente: Stutt-
gart, Ferdinand Enke, xi and 108 pp. "Diese universale Bedeutung der
Geologie muss uns nicht nur anspornen, eifrig an der Losung der vielen
geologischen Probleme weiterzuarbeiten, sondern auch veranlassen, unsere
Forschungsergebmsse in emer verstandlichen und klaren Sprache mitzuteilen, die
jeder Gebildete verstehen kann. Nur so ist es moglich, dass die Geologie der
Menschheit dort dient, wo sie ihr unentbehrlich ist: als Grundstein fur das
gesamte Weltbild der Menschheit"

HABER, FRANCIS COLIN, 1959, The age of the world: Moses to Darwin: Balti-
more, Johns Hopkins Press, xi and 303 pp. "It seems to me that the
constant pressure of the Christian view of historical process on views of
natural process helped to preserve a genetic outlook in terms of concrete,
actualistic time. It was largely a matter of conditioning and prejudice
during the sixteenth, seventeenth and eighteenth centuries, but it held the
potential in readiness until the geologists discovered the real chronology
of the earth in fossil strata ..."

HACK, JOHN TILTON, 1960, Interpretation of erosional topography in humid
temperate regions: Am. J. Sci., vol. 258-A (Bradley Volume), pp. 80-97.
"The theory of dynamic equilibrium explains topographic forms and the
difference between them in a manner that may be said to be independent
of time."

HALDANE, JOHN BURDON SANDERSON, 1944, Radioactivity and the origin of life
in Milne's cosmology: Nature, vol. 153, no. 3888, p. 555. Milne's cos-
mological theory permits description of events in terms of two different

286 CLAUDE C. ALBRITTOX, JR

time scales, which have different implications for geologic history. Use of
a time scale based on a finite past and Euclidean space implies that the
rate at which energy- has been liberated through subatomic or chemical
change has been constant. Use of a scale based on hyperbolic space im-
plies, conversely, that in the geological past chemical change was less
efficient as a source of mechanical energy- than it is now or will be in the
future.

2 1949, Human evolution: past and future. Chapter 22, pp. 405-418, in

Glenn L. Jepsen, Ernst Mayr, and George Gaylord Simpson, eds., Genet-
ics, paleontoiogv, and evolution: Princeton, Princeton Univ. Press, xiv
and 4"4 pp. "We are polymorphic not onlv in our aesthetic but in our
intellectual abilities Ways of describing the world as different as analytical
and projective geometry may be equally true, even if at present one human
mind cannot accept more than one of them at a time."

3 1956, Can a species concept be justified?: Systematics Assoc., London,

Pub. no. 2, pp. 95-96. A species is a name given to a group of organisms
as a matter of convenience and necessity. "... the concept of a species is
a concession to our linguistic habits and neurological mechanisms."

HALL, J^MES, 1883, Contributions to the geological history of the American
continent: Am. Assoc. Adv. Sci., Pr., 1882, vol. 31, pp. 29-69. "The
grand problem of geology- is the entire history, chemical, physical, zoologi-
cal and botanical, of the groups of strata constituting the formations of the
globe."

HAMEISTER, ERNST, 1935, Cbersichtstabelle der Methoden der angewandten
Geophvsik in der praktischen Geologic: Zs. prakt. Geol., vol. 43, no. 2,
pp. 26-29. Shows in tabular form how different geophysical methods
electrical, seismic, magnetic, and gravimetric may be used to investigate
various geological phenomena.

HARRASSOWITZ, HERMANN, 1936, Die Grenzen geologischer Erkenntnis: Forsch.
u. Fortschr., Berlin, vol. 12, no. 28, pp. 355-357. "Bei dieser ontologischen
Methode oder der Anwendung des Grundsatzes des Aktualismus ist
natiirlich nie der Beweis zu erbringen, dass in der geologischen Vergangen-
heit Gesteine nicht auch unter anderen Umstanden entstanden sind."

2 1936, Die Grenzen geologischer Erkenntnis in ihrer Bedeutung fur Geo-
logic und Bergbau: Metall u. Erz, vol. 33, no. 16, pp. 425-431. "Geo-
logische Schliisse geschichtlicher Art bleiben immer nur Annahmen, die
niemals als Tatsachen erweisbar sind."

HARRLNGTON, JOHN WILBUR, and HAZLEWOOD, E. L., 1962, Comparison of
Bahaman land forms with depositional topography of Nena Lucia dune-
reef-knoll, Nolan County, Texas: study in uniformitarianism: Am. Assoc.
Petroleum Geol., B., vol. 46, no. 3, pp. 354-373. The development of

BIBLIOGRAPHY 287

present land forms in the Bahamas is taken as a basis for explaining the
growth of a reef-knoll in the sub-surface Pennsylvanian of the Midland
Basin in Texas.

HAUGHTON, SIDNEY HENRY, 1957, The geophysicist and some geological prob-
lems: Roy. Soc. S. Africa, Tr., vol. 35, pt. 2, pp. 59-69. "Observations are
still the backbone of the science of geology: but to those made by the eye
in the field or in the petrological, mineralogical, and palaeontological
laboratories we now add the observations made by the varied types of
instrument that the physicist has placed at our disposal."

HAWKES, LEONARD, 1957-1958, Some aspects of the progress of geology in the
last fifty years; I: Geol. Soc. London, Quart. J., vol. 113, pp. 309-321,
1957; II: ibid., vol. 114, pp. 395-410, 1958. Uniformitarians, in the
Lyellian sense, are back again in force, although it is not yet clear how
atmospheric oxygen or Precambrian jaspillite banded ironstone cam be
explained according to the dictum that the present is the key to the past.

HAWKINS, HERBERT LEADER, 1936, Palaeontology and humanity: British Assoc.
Adv. Sci., Rept., 1936, pp. 57-80. "Just as a net has been described as a
set of holes held together with string, so a series of strata must often repre-
sent a succession of non-sequences separated by films of sediment."

2 1938, Humanity in geological perspective: British Assoc. Adv. Sci., Rept.,

1938, pp. 546-556. "The hoary imposture of the accuracy of the 'exact'
sciences still deludes mankind, through the wildly illogical belief that a
rigidly logical argument must reach a correct result whatever errors may
have existed in the premisses on which it is based."

HEDBERG, HOLLIS Dow, 1948, Time-stratigraphic classification of sedimentary
rocks: Geol. Soc. Am., B., vol. 59, no. 5, pp. 447-462. ". . . man's . . . need
to apply an arbitrary and strictly delimited classificatory system to what
in Nature is frequently a continuous process or gradual change in charac-
ters results from the clamoring of the human mind for steps on which to
rest as a relief from the more precarious footing of the continuous slope
which often more correctly represents these natural relations."

2 1959, Towards harmony in stratigraphic classification: Am. J. Sci., vol.

257, no. 10, pp. 674-683. "Stratigraphic classification is the systematic
zonation of the strata of the earth's crust with reference to any of the
properties or attributes which rock strata may possess."

HEMPEL, CARL GUST A v, 1942, The function of general laws in history: J. Phil.,
vol. 39, no. 1, pp. 35-48. "In history as anywhere else in empirical science,
the explanation of a phenomenon consists in subsuming it under general
empirical laws; and the criterion of its soundness is not whether it appeals
to our imagination, whether it is presented in suggestive analogies, or is

288 CLAUDE C. ALBRITTOX, JR.

otherwise made to appear plausible . . . but exclusively \vhether it rests
on empirically well confirmed assumptions concerning initial conditions
and general laws."

2 1958, The theoretician's dilemma, a study in the logic of theory construc-
tion; ZTZ H. Feigl, M. Scriven and G. Maxwell, eds., Minnesota Studies in
the Philosophy of Science, vol. 2: Minneapolis, Univ. Minnesota Press,
pp. 37-98. "Scientific explanation, prediction, and postdiction all have
the same logical character: they show that the fact under consideration
can be inferred from certain other facts by means of specified general
laws."

HENBEST, LLOYD GEORGE, 1952, Significance of evolutionary explosions for
diastrophic division of earth history (Introduction to a symposium on the
distribution of evolutionary explosions in geologic time): J. Paleont.,
vol. 26, pp. 299-318. "Without intending to deny that rhythm or vibra-
tion is a common characteristic of natural processes, it is evident that the
theorizing on rhythm in nature has outrun the facts and possibly its
actual importance."

HERSCHEL, Sir JOHN FREDERICK WILLIAM, 1831, A preliminary discourse on the
study of natural philosophy, in The cabinet of natural philosophy con-
ducted by the Rev. Dionysius Lardner: Philadelphia, Carey and Lea,
279 pp. "... geologists have no longer recourse, as formerly, to causes
purely hypothetical . . . ; but rather endeavor to confine themselves to a
consideration of causes evidently in action at present, with a view to ascer-
tain how far they, in the first instance, are capable of accounting for the
facts observed, and thus legitimately bringing into view, as residual
phenomena, those effects which cannot be so accounted for."

2 1841, Whewell on inductive sciences: Quart. Rev. (London), vol. 68, pp.

177-238. "The most strenuous advocate for the exclusion of paroxysmal
epochs will not contend for perfect uniformity so long as earthquakes are
not of daily occurrence . . . : and the question as to what is and what is
not paroxysm, to what extent the excursion from repose or gentle oscilla-
tion may go without incurring the epithet of a catastrophe, is one of mere
degree, and of no scientific importance whatsoever."

HOLDER, HELMUT, 1950, Geologic und Palaontologie in Texten und ihre
Geschichte: Munich, Karl Alber Freiburg, xviii and 566 pp. A source
book for geology- and paleontology. Part 1, the last three sections of part 3,
and part 4 are given largely to philosophical citations. Lengthy bibli-
ography.

HOFF, KARL ERNST ADOLF VON, 1822-1841, Geschichte der durch Uberlieferung
nachgewiesenen natiirlichen Veranderungen der Erdoberflache: Gotha,
Justus Perthes (vols. 1-3), H. Berghaus (vols. 4-5). ". . . iiberwiegende

BIBLIOGRAPHY 289

Griinde erlauben nicht nur, sondern fordern sogar, dass man die Veran-
derungen, die man auf der Erdoberflache wahrgenommen hat, und noch
wahrnimmt, nicht nur als auf einzelne Theile und Gegenden derselben
beschrankt betrachten muss, sondern auch dass man sie keinen ausseror-
dentlichen Naturwirkungen, welche aufgehort haben, sondern allein der
Wirkung derjenigen Krafte zuschreiben darf, durch die man noch jetzt
alle und jede Naturerscheinungen hervorgebracht sieht; und dass die fur
uns unermessliche Grosse der Zeitraume, in welchen diese Krafte allmah-
lich und immerfort gewirkt haben, geniigt, die Veranderungen durch eben
diese Krafte hervorbringen zu lassen." (v. 3, pp. 252, 1834)

HOOYKAAS, REIJER, 1956, The principle of uniformity in geology, biology and
theology: Victoria Inst., J. of Tr., vol. 88, pp. 101-116. "Strict uniformi-
tarianism may often be a guarantee against pseudo-scientific phantasies
and loose conjectures, but it makes one easily forget that uniformity is not
a law, not a rule established after comparison of facts, but a methodological
principle, preceding the observation of facts. It is the logical principle of
parsimony of causes and of economy of scientific notions."

2 1959, Natural law and divine miracle: a historical-critical study of the

principle of uniformity in geology, biology and theology: Leiden, E. J.
Brill, 237 pp. "A less strict maintenance of uniformitarianism would
perhaps give freer play to fantasy, but it might also open new vistas. The
uniformitarian position, at its worst, forces past phenomena into a pre-
conceived frame built upon events occurring in our time."

HORBERG, CARL LELAND, 1952, Interrelations of geomorphology, glacial geol-
ogy, and Pleistocene geology: J. GeoL, vol. 60, no. 2, pp. 187-190. "... a
central aspect of geomorphology is its application to the task of deciphering
Pleistocene history."

HORTON, ROBERT ELMER, 1945, Erosional development of streams and their
drainage basins, hydrophysical approach to quantitative morphology:
Geol. Soc. Am., B., vol. 56, no. 3, pp. 275-370. Develops laws of drainage
composition, law of stream slopes, and law of overland flow.

HUBBERT, MARION KING, 1937, Theory of scale models as applied to the study
of geologic structures: Geol. Soc. Am., B., vol. 48, no. 10, pp. 1459-1520.
A general theory of the similarity between a model and its original, for
purely mechanical systems, is derived and then applied to illustrative
geologic problems.

2 1938, The place of geophysics in a department of geology: Am. Inst. Min.

Met. Eng., Tech. Pub. 945, 19 pp. Ranked according to degree of de-
pendence on other sciences for fundamental concepts, geology is a third-
order discipline subordinate to physics, chemistry, and astronomy. Geo-
physics and geochemistry are not sciences, properly speaking, but are

290 CLAUDE C. ALBRITTON, JR.

rather the vehicles by which a more dependent science incorporates the
data and techniques of a less dependent science.

3 (Hendricks, Thomas Andrews and Thiel, George Alfred, Chm.), 1949 Re*

port of the Committee on Geologic Education of the Geological Society of
America: Geol. Soc. Am., Int. Pr., 1949, pt. 2, pp. 17-21. ". . .your
Committee recommend that at all instructional levels from the most elementary
to the most advanced, only those inferences be presented to students for which the
essential observational data and the logical steps leading to the inference have also
been presented. The satisfaction of this criterion will compel a badly needed
critical reexamination from the ground up of the logical structure of geo-
logical science. 5 '

HUNT, CHARLES B., 1956, A skeptic's view of radiocarbon dates: Univ. of Utah,
Anthropological Papers, no. 26, pp. 35-46. "Consistency of results, in-
cluding consistency in a series of radiocarbon dates, may be evidence of
precision, but it is not evidence of accuracy although it commonly is cited
as such."

MUTTON, JAMES, 1788, Theory of the earth; or an investigation of the laws
observable in the composition, dissolution, and restoration of land upon
the globe: Roy. Soc. Edinburgh, Tr., vol. 1, pp. 209-304. "In examining
things present, we have data from which to reason with regard to what has
been; and from what has actually been, we have data for concluding with
regard to that which is to happen hereafter. Therefore upon the supposi-
tion that the operations of nature are equable and steady, we find, in
natural appearances, means for concluding a certain portion of time to
have necessarily elapsed, in the production of those events of which we see
the effects."

HUXLEY, THOMAS HENRY, 1862, The anniversary address to the Geological
Society: Geol. Soc. London, Quart. J., vol. 18, pp. xl-liv. "Allied with
geology-, palaeontology has established two laws of inestimable importance:
the first, that one and the same area of the earth's surface has been succes-
sively occupied by very different kinds of living beings; the second, that
the order of succession established in one locality holds good, approxi-
mately, in all."

2 1881, The rise and progress of palaeontology- Nature, vol. 24, pp. 452-455.

"The whole fabric of palaeontology' is based upon two propositions: the
first is, that fossils are the remains of animals and plants; and the second
is, that the stratified rocks in .which they are found are sedimentary de-
posits; and each of these propositions is founded upon the same axiom
that like effects imply like causes."

Interdepartmental Stratigraphic Committee, USSR, 1956, Stratigraphic classi-
fication and terminology (Translated from the Russian by John Rodgers.):
Int. Geol. Rev., vol. 1, Feb. 1959, pp. 22-38. A sfcheme of classification is

BIBLIOGRAPHY 291

developed on the premises that stratigraphic subdivisions exist objectively
in Nature, rather than subjectively in the mind of the investigator, and
that these subdivisions reveal the actual course of geologic history.

JACOBS, JOHN ARTHUR, and ALLAN, D. W., 1956, The thermal history of the
earth: Nature, vol. 177, no. 4500, pp. 155-157. Studies of earth models
suggest that the crust of the earth, down to about 100 km, may have been
remelted during the first billion or so years of earth history, that the rate
of cooling of the earth was greater in the past than now, and thus that
orogenic activity may have decreased with time.

JAMES, HAROLD LLOYD, 1960, Problems of stratigraphy and correlation of Pre-
cambrian rocks with particular reference to the Lake Superior region:
Am. J. Sci., vol. 258-A (Bradley Volume), pp. 104-114. "The most basic
geologic law is that of superposition. Though the principle itself can be
called obvious, its application to highly deformed rocks is rarely easy."

JELETZKY, JURIJ ALEXANDER, 1956, Paleontology, basis of practical geochronol-
ogy: Am. Assoc. Petroleum Geol., B., vol. 40, no. 4, pp. 679-706.

2 1962, The allegedly Danian dinosaur-bearing rocks of the globe and the

problem of the Mesozoic-Cenozoic boundary: J. Paleont., vol. 36, no. 5,
pp. 1005-1018. "This obviously geologically contemporary extinction of
the terrestrial and marine Mesozoic animals at the Maestrichtian-Danian
boundary must reflect some kind of radical, world-wide change in the
physical regime of our planet. This event can be quite properly referred
to as a 'catastrophe' or 'revolution 9 . . ."

JENNINGS, HERBERT SPENCER, 1933, The universe and life: New Haven, Yale
Univ. Press, 94 pp. "The universe would be so much easier to deal with
if the laws of its action were simple and uniform. Let us assume therefore
that they are simple and uniform ! The doctrine is a marked case of wish-
ful thinking."

JEPSEN, GLENN LOWELL, 1949, Selection, "orthogenesis" and the fossil record:
Am. Phil. Soc., Pr., vol. 93, no. 6, pp. 479-500. "Orthogenesis was fre-
quently expressed as a e law s at a time when there was an eager search for
biological phenomena or principles which could be called laws, after the
model of the eponymic laws of physics and chemistry."

JOHNSON, DOUGLAS WILSON, 1933, The r&le of analysis in scientific investiga-
tion: Geol. Soc. Am., B., vol. 44, no. 3, pp. 461-494. "Multiple working
hypotheses as a method, employed in connection with critical analysis
as an instrument of precision, offer us ... the best guarantee of success in
scientific research."

292 CLAUDE C. ALBRITTOX, JR.

2 1938-1942, Studies in scientific method- J. Geomorphology, vol. 1, pp.

64-66, 147-152; vol. 2. pp. 366-372; vol. 3, pp. 59-64, 156-162, 256-262,
353-355: vol. 4, pp. 145-149, 328-332; vol. 5, pp. 73-77, Tl-173. "Hence
the golden rule of scientific exposition: \eicr discuss an unknown in teims of
an unknown."

JOHNSON, MARTIN, 1951, The meanings of time and space in philosophies of
science: Am. Scientist, vol. 39, no. 3, pp. 412-421. "Crudely stated, it
may be said that physical science is interested in the changing or the flux
of the world, not in any static picture, and is in fact a study of a sequence
of events whose basic pattern is a time-order/'

JORALEMON, IRA SEAMAN, 1952, Age cannot wither, or varieties of geological
experience: Econ. Geol., vol. 47, no. 3, pp. 243-259. ". . . we have not
sufficient reliable evidence ... for hard-and-fast generalizations or laws of
ore occurrence/ 5

JOYSEY, KENNETH A., 1952, Fossil lineages and environmental change: Geol.
Mag., vol. 89, no. 5, pp. 357-360. "An essential key to the interpretation
of fossil communities lies in the study of living animals on the present day
time-plane. Equally, the process of evolution can only be seen in perspec-
tive when the zoological picture is projected along the dimension of geo-
logical time."

JUDSON, SHELDON, 1958, Geomorphology and geology: New York Acad. ScL,
Tr., ser. 2, vol. 20, no. 4, pp. 305-315. "Geology is history. Perhaps more
than any other single characteristic, the time factor distinguishes geology
from the other sciences. 35

2 1960, William Morris Davis an appraisal: Zs. Geomorph., vol. 4, no. 3-4,

pp. 193-201. "The cycle of erosion had an almost immediate acceptance
not only because of its skilful presentation but also because the geologic
profession was ready for such a synthesis. Uniformitarianism, a major
element in the development of the cycle, was by then almost universally
accepted by geologists. Furthermore, the orderly development of land-
forms through successive stages represented a type of non-organic evolu-
tion ... in harmony with the exciting new ideas in organic evolution then
sweeping the scientific world. 53

3 1961, Archaeology- and the natural sciences: Am. Scientist, vol. 49, no. 3,

pp. 410-414. "The archaeologist reconstructs history largely from objects
recovered from the ground . . . Such reconstructions are generally classed
under the heading of c pre-history% a term that has always bothered me a
bit because the inference exists that events of the pre-literate period do not
combine to constitute history, a particularly unacceptable inference to a
geologist."

BIBLIOGRAPHY 293

KAISER, ERICH, 1931, Der Grundsatz des Aktualismus in der Geologic: Deut.
Geol. Ges,, Zs., vol. 83, pp. 389-407. "Aber ganz abwegig ist es, wenn
man jetzt mehr und mehr von einem Gesetz des Aktualismus liest, trotzdem
doch die Begriinder und Verfechter des Aktualismus nur von einer ak-
tualistischen (ontogenetischen) Methode redeten."

KALKOWSKY, ERNST, 1910, Geologic und Phantasie: Naturw. Ges. Isis, Dresden,
Sitzungsber. u. Abh., Jan-June, 1910, pp. 10-19. "Nie hat jemand cine
ganze Schicht gesehen, immer nur Stuckchen davon hat er vor Augen
gehabt, und das Ganze ist in jedem einzelnen Falle nichts als ein Phan-
tasiegebilde. Der Begriff der Schicht, meine ich, steht somit etwa auf
derselben Stufe, wie der des Atoms . . /'

KELVIN, WILLIAM THOMSON, Baron, 1864, On the secular cooling of the earth:
Roy. Soc. Edinburgh, Tr., vol. 23, pt. 1 (1861-1862), pp. 157-169. "It
must be admitted that many geological writers of the 'Uniformitarian'
school, who in other respects have taken a profoundly philosophical view
of their subject, have argued in a most fallacious manner against hypotheses
of violent action in past ages. If they had contented themselves with show-
ing that many existing appearances, though suggestive of extreme violence
and sudden change, may have been brought about by long-continued
action, or by paroxysms not more intense than some of which we have
experience . . . their position might have been unassailable ..."

2 1871, On geological time: Geol. Soc. Glasgow, Tr., vol. 3, pt. 1, pp. 1-28.

"It would be just as reasonable to take a hot water jar, such as is used in
carriages, and say that that bottle has been as it is for ever as it was for
Playfair to assert that the earth could have been for ever as it is now, and
that it shows no traces of a beginning, no progress towards an end."

KEMENY, JOHN GEORGE, 1959, A philosopher looks at science: Princeton,
D. Van Nostrand, xii and 273 pp. The scientist's assumption of the "uni-
formity of nature" simply asserts that human capacities and the complexity
of nature are such that it is possible for scientists to learn something about
nature in a reasonably short time. See pp. 59-64 for a discussion of the
principle of uniformity.

KERMACK, K. A., 1956, Species and mutations: Systematics Assoc., London,
Pub. no. 2, pp. 101-103. "Our justification for creating species, muta-
tions, etc., in paleontology is solely one of practical convenience: the cate-
gories themselves are man-made and artificial."

KHAIN, VIKTOR EFIMOVICH, 1958, Nekotorye filosofskiye voprosy sovremennoy
geologii (Some philosophic questions of modern geology): Nauchnye
Doklady Vysshei Shkoly, Filosofskiye Nauki, vol. 2; Moskva, pp. 148-161.
Soviet scientists approach each theoretical question from the standpoint

294 CLAUDE C. ALBRITTON. JR.

of dialectic materialism; each geological concept in the Soviet science
follows the idea of a directed and irreversible development.

2 1961, Lomonosov i sovremennaya geologiya (Lomonosov and modern

geology-): Sovet. Geol., vol. 12, pp. 14-28. Since the end of World War II,
geology* has tended to change from a descriptive to an explanatory science
whose aim is to discover the basic laws of the development of the planet
Earth, and of the earth's crust in particular.

KING, CLARENCE, 1877, Catastrophism and evolution: Am. Naturalist, vol. 11,
pp. 449-470. "Men are born either catastrophists or uniformitarians.
You may divide the race into imaginative people who believe in all sorts
of impending crises physical, social, political and others who anchor
their very souls in status quo "

KING, LESTER CHARLES, 1953, Canons of landscape evolution: Geol. Soc. Am.,
B., vol. 64, no. 7, pp. 721-752. * w There is a general homology between all
epigene landscapes."

2 1957, The uniformitarian nature of hillslopes: Edinburgh Geol. Soc., Tr.,

vol. 17, pt. 1, pp. 81-101. ". . . the manner of hillslope evolution is essen-
tially uniformitarian in all climatic realms outside the frigid zones and the
erg deserts."

KITTS, DAVID B., 1963, Historical explanation in geology: J. Geol., vol. 71,
pp. 297-313.

KOBAYASHI, TEIIGHI, 1944, Concept of time in geology: Imp. Acad. Tokyo,
Pr., vol. 20, no. 7, pp. 475-478 (On the major classification of the geologis
cal age.); no. 10, pp. 742-750 (An instant in the Phanerozoic Eon and it-
bearing on geology and biology.); pts. 1 and 3 in a series of articles.
"Time . . . may be classed into two major kinds, one with and the other
without historical contents. Time of the latter kind is abstract, absolute,
objective and physical; that of the former kind, concrete, relative, sub-
jective and historical. Geological age belongs to the former while chronology
is of the latter kind . . ."

KOBER, LEOPOLD, 1946, Geo-Logismus: Geog. Ges. Wien, Mitt. vol. 89, no. 1-6,
pp. 3-7. "Absolute kosmo-geo-logische Wahrheit aller Evolution der Erde
ist der Geo-Logismus: die Vergeistigung der Natur in Menschen. Der
Mensch ist das Mittel der Natur, sich selbst zu erkennen."

KOCH, LEO E., 1949, Tetraktys; the system of the categories of natural science
and its application to the geological sciences: Australian J. Sci., vol. 11,
no. 4, suppl., 31 pp. "This method of application of the Tetraktys coin-
cides with Aristotle's doctrine . . . that a system of categories of a very
elementary nature is the necessary basis of any scientific definition."

BIBLIOGRAPHY 295

KRUMBEIN, WILLIAM CHRISTIAN, 1960, The "geological population" as a frame-
work for analysing numerical data in geology: Liverpool and Manchester
Geol. J., vol. 2, pt. 3, pp. 341-368. Statistical analysis may be applied to
classes and aggregates of geological objects and events, toward the ends of
deriving inferences and making predictions. (With annotated bibliog-
raphy.)

2 1962, The computer in geology; quantification and the advent of the com-
puter open new vistas in a science traditionally qualitative: Science, vol.
136, no. 3522, pp. 1087-1092. "The present dominantly empirical aspect
of much data analysis in geology is not disturbing in a science where much
effort . . . must still be directed toward a search for controls and responses
in a web of intricately interlocked data."

KRYNINE, PAUL DIMITRI, 1951, A critique of geotectonic elements: Am. Geo-
phys. Union, Tr., vol. 32, no. 5, pp. 743-748. ". . . it seems that the
classical principle of the description and classification of landforms as
introduced by W. M. Davis the concepts of describing landforms in
terms of structure, process, and stage could be fruitfully applied to the
classification of geotectonic elements. This powerful generalization of the
immortal Davis is indeed the great c universal solvent' of geology, and no
subject in the Earth sciences contains problems that would not yield to
this method, providing that adequate data can be gathered."

KUENEN, PHILIP HENRY, 1958, Experiments in geology: Geol. Soc. Glasgow,
Tr., vol. 23, pp. 1-28. "In a certain sense present day processes show an
experimental relation to historical geology and the investigation of ancient
rocks . . . Hence . . . the 'Present as key to the Past' is an experimentally
tainted maxim."

KULP, JOHN LAURENCE, 1961, Geologic time scale; Isotopic age determinations
on rocks of known stratigraphic age define an absolute time scale for earth
history: Science, vol. 133, no. 3459, pp. 1105-1114.

KUMMEROW, EGMONT, 1932, Die aktualistische Methode in der Geologic: Deut.
Geol. Ges., Zs., vol. 84, pp. 563-565. "Der Grundsatz des Aktualismus
besitzt keine uneingeschrankte Gultigkeit. Er lasst sich auf periodische
Vorgange im allgemeinen nicht anwenden. Die Gegenwart ist um so
weniger geeignet, fur alle Krafte und Vorgange der Vergangenheit Bei-
spiele zu liefern, als sie keine normale Warmzeit, sondern cine glaziale
Ausnahmezeit darstellt."

LADD, HARRY STEPHEN, 1957, Introduction, and Paleoecological evidence,
Chapters 1 and 2, pp. 1-66, in Treatise on marine ecology and paleoecol-
ogy, vol. 2, Paleoecology (H. S. Ladd, ed.): Geol. Soc. Am., Mem. 67.
Describes the nature of the paleontological record, and explains how

296 CLAUDE C. ALBRITTON, JR.

ancient marine environments are reconstructed on the basis of paleonto-
logical and lithological evidence.

LAHEE, FREDERIC HENRY, 1909, Theory and hypothesis in geology: Science,
n.s., vol. 30, pp. 562-563. "The misconception of the need for unity of
cause may be an outgrowth from the doctrine of uniformity. But uni-
formity is not synonymous with simplicity, any more than complexity is
synonymous with chaos."

LAIRD, JOHN, 1919, The law of parsimony: Monist, vol. 19, no. 3, pp. 321-344.
"The complexity of the universe can never be simplified out of existence . . .
It is conceivable, therefore, that the law of economy is neither more nor
less than a rule of direction stating that we should always select the simplest
of any general propositions which may be true and ascertainable, since it
is these truths only that are likely to aid the mind in making further dis-
coveries. This rule shows the path of wisdom in view of the limited powers
and scope of the mind."

LAKE, PHILIP, 1930, The centenary of LyelTs Principles of Geology: Geol.
Mag., vol. 67, pp. 433-436. "Each attitude of mind has its own dangers,
and the chief danger of the uniformitarian attitude is that the mind may
be closed to new ideas . . . There is little need now, however, to dwell on
the dangers of the uniformitarian attitude. There is nothing uniformitarian
about post-war geological theories, and there are so many of them that it
is difficult not to keep an open mind. Meanwhile geology progresses in
spite of theories."

LAMONT, ARCHIE, 1944-1945, Geology in literature: Quarry Managers' J.,
vol. 27, pp. 555-560, 567; vol. 28, pp. 28-31, 77-79, 121-127, 194-202,
287-291, 365-369, 405-407, 439-441, 495-499, 551-554. Illustrates, with
many quotations, the impact of geological ideas on literature. "Geology
is par excellence the science which deals with the repetition of the same events
in widely separated epochs."

LANE, ALFRED CHURCH, 1906, The geologic day: J. Geol., vol. 14, pp. 425-429.
"Time is measured by change, and change . . . must progress from point
to point . . . there is always a rate of progress which is measurable, and
any interval of time must be marked by a certain stage in the change, and
so begin differently at different points."

LASKY, SAMUEL GROSSMAN, 1947, The search for concealed deposits a re-
orientation of philosophy: Am. Inst. Min. Met. Eng., Tech. Pub. 2146,
Mining Tech., vol. 11, no. 3, 8 pp. "Instead of looking for information
indicating that ore is likely to be present, we ought ... to balance and fit

BIBLIOGRAPHY 297

together the observed facts so as to discover whether [it] could be present.
In other words, the geology need be only permissive instead of definitely
indicative."

LAUNAY, Louis DE, 1922, La science g6ologique, ses m6thodes, ses r&ultats, ses
problfcmes, son histoire: Paris, Armand Colin, 3d. ed., viii and 775 pp.
A survey of geological methodology and the evolution of geological ideas.

LAUTERBACH, ROBERT, 1957, Palaogeophysik, in Geol. u. Geophys. (Karl-
Marx-Univ., Geol. u. Pal. Inst.), pp. 57-66.

LECLERQ,, REN, 1960, Histoire et avenir de la mithode exp&imentale: Paris,
Masson, 138 pp. "La recherche des puits de p6trole par Pexamen g6o-
logique du sol est devenue courante. Ce n'est qu'un cas particulier de la
m6thode qui consiste pr6voir une allure de stratification ou une couche
d6termin6e un endroit pr6cis i partir de la connaissance de la gfeologie
d'une region. On 6met une hypothfese en se fondant sur des observations
et on la v6rifie par une autre observation."

LE CONTE, JOSEPH, 1900, A century of geology: Pop. Sci. Monthly, vol. 56,
pp. 431-443, 546-556. "The fundamental idea underlying geological
thought is the history of the earth."

LE GRAND, HARRY E., 1962, Perspective on problems of hydrogeology: Geol.
Soc. Am., B., vol. 73, pp. 1147-1152. "Mathematical research relating to
synthesis of the elements of hydrogeology is showing some promise, but it
tends to wilt when faced with a kaleidoscope of hydrologic conditions
caused by heterogeneities of geology."

LENZEN, VICTOR F., 1955, Procedures of empirical science: Int. Encyc. Unified
Sci., vol. 1, pt. 1, Chicago, Univ. Chicago Press, pp. 280-339. "A stage
in the investigations of correlations of events is the determination of tem-
poral sequences. The history of political events, historical geology, and
paleontology are arrangements of events in temporal order. Science in
the form of history systematizes observations of events by fitting them into
schemes of development of the cosmos, life and society."

LEOPOLD, LUNA B., and LANGBEIN, WALTER B., 1962, The concept of entropy
in landscape evolution: U. S. Geol. Survey, Prof. Paper 500-A, 20 pp.
". . . we believe that the concept of entropy and the most probable state
provides a basic mathematical conception which does deal with relations
of time and space. Its elaboration may provide a tool by which the various
philosophic premises still characterizing geomorphology may be subjected
to critical test."

298 CLAUDE C. ALBRITTON, JR.

LINTON, DAVID LESLIE, 1948, The ideal geological map: Adv. Sci., vol. 5, no.
17, pp. 141-149. The symbols of geological cartography will require
standardization on an international scale before all geological maps will
be intelligible to all geologists. Choices of color symbols have been vari-
ously determined by the "principle of the distinctive horizon" (bold colors
for index formations), by the association of ideas (fiery red for volcanic
rocks), by the "principle of associative colouring" (e.g., red for the Old
Red sandstone), and by reference to a spectrum arbitrarily selected to
indicate relative ages of rocks. Even greater diversity and confusion exist
in the array of index letters and numbers variously sequential, nonse-
quential, or hybrid used to designate map units.

LOEWINSON-LESSING, FRANTS YULEVICH, 1954, A historical survey of petrology
(Translated from the Russian by S. I. Tomkeieff): London, Oliver and
Boyd, vi and 112 pp. "There is one fundamental and leading principle
which lies at the base of petrological science. This principle was dimly
perceived at the dawn of modern geology and petrology and since that
time its conception has been steadily growing and expanding. The prin-
ciple is that of development the evolution or gradual change to which
every rock is subjected from the very moment of its formation."

LOSSKIY, NIKOLAI ONUFRIEVICH, 1956, Organicheskii aktualizm Bergsona
(Organic actualism of Bergson) in An introduction to philosophy, Chapter
XX, pp. 190-198: Frankfurt a. M., Possev. Actualism is a philosophic
concept according to which the whole world consists of events and proc-
esses which appear and disappear in time.

Lucius, MICHEL, 1957, Les principes fondamentaux de travail et de recherche
en g6ologie: Soc. Nat. Luxembourg, B., n.s., vol. 49, no. 60, pp. 141-143.
"Mais le principe d'actualisme n'est pas toujours apparent. Nous ne
voyons pas se former aujourd'hui du granite, des schistes cristallins. Dans
la nature, les deux 6tats extremes d'un mme ph6nomfcne peuvent &tre si
diff6rents, qu'il n'apparait, a premiere vue, aucune connexion entre eux.
Mais si on trouve une suite d'6tats interm6diaires entre les deux extremes,
on peut admettre un 6tat original identique. C'est le principe des flats
intermediates"

LYELL, CHARLES, 1830, Principles of geology, being an attempt to explain the
former changes of the earth's surface, by reference to causes now in opera-
tion, vol. 1: London, John Murray, 511 pp. "Our estimate ... of the
value of all geological evidence . . . must depend entirely on the degree of
confidence which we feel in regard to the permanency of the laws of nature.
Their immutable constancy alone can enable us to reason from analogy,
by the strict rules of induction, respecting the events of former ages."

BIBLIOGRAPHY 299

MAGGREGOR, ALEXANDER MIERS, 1951, Some milestones in the Precambrian
of Southern Rhodesia: Geol. Soc. S. Africa, Pr., vol. 54, pp. xxvii-lxvi.
"In dealing with rocks formed when the world was less than half of its
present age, a strict adherence to the doctrine of uniformitarianism is
considered unjustified . . . the [older Precambrian] sedimentary rocks were
closer in their mineral composition to igneous rocks and were therefore
more amenable to granitization. The geothermal gradient was probably
steeper . . . The crust of the earth may have been less rigid . . . Volcanism
was more active than it is now . . . The earth's rotation was more rapid
and tidal forces may have been greater."

MCLACHLAN, DAN, JR., 1961, A guess as to what is science: Physics Today,
vol. 14, no. 6, pp. 22-27. "When one science swallows another ... the
general attitude among scientists is that the new science has built a firm
foundation under the old science. The physicists and chemists would like
to do this for geology . . ."

MCLAREN, DIGBY JOHNS, 1959, The role of fossils in defining rock units, with
examples from the Devonian of western and Arctic Canada: Am. J. Sci.,
vol. 257, no. 10, pp. 734-751. "All geology depends on one basic prin-
ciple that it is possible to interpret the history of the earth by examining
the positional relationships of rock and mineral bodies. Stratigraphy relies
primarily on a special case of this principle the positional relationships
of stratified rock bodies, from which derives the law of superposition.

MANLEY, GORDON, 1953, Climatic variation: Roy. Meteorological Soc., Quart.
J., vol. 79, pp. 185-209. (Reviews of modern meteorology No. 9) "The
forecaster on a November day who considers whether a low moving up
the Baltic will occlude more or less rapidly on account of a newly estab-
lished snow cover in Lithuania is probably touching on the whole problem
of the alimentation of the Pleistocene ice caps. But the time scale is very
different. Yet it is the present writer's conviction that in the present can
be sought the key to the past; the day-to-day synoptican will do well to
integrate some of his findings into a greater pattern."

MARSH, OTHNIEL CHARLES, 1879, The history and methods of palaeontological
discovery: Am. Assoc. Adv. Sci., Pr., vol. 28, no. 1880, pp. 1-42. "If I
may venture ... to characterize the present period in all departments of
science, its main feature would be a belief in universal laws"

MASON, STEPHEN FINNEY, 1953, Main currents of scientific thought: New York,
Henry Schuman, viii and 520 pp.

MAYR, ERNST, 1951, Bearing of some biological data on geology: Geol. Soc.
Am., B., vol. 62, no. 5, pp. 537-546. "As far as paleontology is concerned,

300 CLAUDE C. ALBRITTON, JR.

there has been a rather futile argument in recent years as to whether it is
a branch of geology* or of biology. The answer, of course, is that it is either
both or neither."

2 1957, Species concepts and definitions, in The species problem. A sym-
posium presented at the Atlanta meeting of the American Association of
the Advancement of Science, Dec. 28-29, 1955: Washington, D. C., Am.
Assoc. Adv. Sci., Pub. 50, pp. 1-22. ". . . the analysis of the species prob-
lem would be considerably advanced, if we could penetrate through such
empirical terms as phenotypic, morphological, genetic, phylogenetic, or
biological, to the underlying philosophical concepts. A deep, and perhaps
widening gulf has existed in recent decades between philosophy and
empirical biology. It seems that the species problem is a topic where
productive collaboration between the two fields is possible."

MELTON, MARK ALDRIDGE, 1958, Correlation structure of morphometric
properties of drainage systems and their controlling agents: J. Geol.,
vol. 66, no. 4, pp. 442-460. "The variability in any natural environment
is the product of the happenings in many geologic periods ... To argue
that this variability could ever be entirely explained is absurd. Geology
differs from physics and other exact sciences in the kind of variability
encountered, as well as the amount ... the concept of the universe ap-
plicable to geology and physics is accordingly quite different; . . . there is
not just a single scientific 'world view 3 , but two or perhaps many."

MERRIAM, JOHN CAMPBELL, 1921, The earth sciences as the background of
history: Sci. Monthly, vol. 12, no. 1, pp. 5-7. "Our greatest scientific
contributions to the study of history and of origins have come through
geological and biological investigations. Geology is the greatest of his-
torical sciences."

MILLER, BENJAMIN LERov, 1941, Geology and the allied sciences: Pennsylvania
Acad. Sci., Pr., vol. 15, pp. 82-89. ". . . the principal reasons for geological
investigation should be economic."

MILLER, HUGH, 1939, History and science; a study of the relation of historical
and theoretical knowledge: Berkeley, Univ. of California Press, x and
201 pp. "Thus nature is still conceived as pure history, in human history,
and as pure structure, in physicochemical theory; and biology and geology
remain alone in the synthesis of historical and theoretical principles. While
everyone speaks of universal evolution, the presence of evolutionary charac-
ter is really admitted only in some branches of biological and geological
science; and even here the dominant intention is often a reduction of his-
torical change to structural law, just as if Darwin had never lived and
taught."

BIBLIOGRAPHY 301

MILLER, JOHN PRESTON, 1959, Geomorphology in North America: Przeglad
Geograficzny, Warsaw, vol. 31, no. 3-4, pp. 567-587. "Like all the other
phases of historical geology, geomorphic history is based ultimately on the
Principle of Uniformitarianism. This view that 'the present is the key to
the past' implies only that physical and chemical mechanisms have been
constant in kind, but it recognizes that their magnitudes and rates may
have varied. In any case, interpretation of history assumes an adequate
knowledge of the present. Contrary to the belief of Davis and his followers
that process was known or could be inferred, current opinion holds that
our knowledge of present processes and their products is so meager that
inferences about geomorphic history rest too heavily on untested assump-
tions."

MILLHAUSER, MILTON, 1954, The scriptural geologists, an episode in the history
of opinion: Osiris, vol. 11, pp. 65-86. "Stratigraphy and paleontology not
only established the preconditions for a biological theory like Darwin's;
in their day they were nearly as revolutionary in their impact on the
popular mind."

MILNE, EDWARD ARTHUR, 1935, Some points in the philosophy of physics:
time, evolution and creation: Smithsonian Inst., Ann. Rept., 1933, pp.
219-238. "The world is thus a continuing system; each particle or nebula
has an evolutionary experience behind it and in front of it, with ultimate
decay as its goal, yet the world as a whole cannot be said to decay. It is
not the same in an observer's today as in the observer's yesterday, but it
is the same forever."

MISES, RICHARD VON, 1951, Positivism, a study in human understanding:
Cambridge, Harvard Univ. Press, vi and 404 pp. "... the various
sciences are distinguished by the objects they deal with ... If we examine
a little more closely what the individual scientists make concrete statements
about, we find that the objects of the various sciences can be summarized
under a common expression: every proposition of a positive science refers
to actions of men and observed by men (Bridgman) . . . The purely
descriptive natural sciences do not constitute an exception . . . even the
geologist can only say what happens if one examines rocks by means of
hammer and drill."

MITCHELL, RAOUL C., 1955, Changement de cadence des ph&aomfenes gfo-
logiques: Cahiers Gol., Seyssel, no. 33, pp. 331-334. The principle of
uniformity must be emended to allow for variations in the rate at which
geologic processes have acted during the past.

MOORE, RAYMOND CECIL, 1948, Stratigraphical paleontology: Geol. Soc. Am.,
B., vol. 59, no. 4, pp. 301-326. ". . . Stratigraphical paleontology is syn-

302 CLAUDE C. ALBRITTON, JR.

onomous with paleontology* itself, viewed with the eyes of a geologist,
whereas so-called systematic paleontology- comprises a purely biological
approach, without reference to geology except for arrangement of fossils
in order of time succession."

2 1949, Meaning of facies: Geol. Soc. Am., Mem. 39, pp. 1-34. ". . . many

features of geological history* can never be learned, because they never
became incorporated in the rock record. These also are actual parts of
earth history, which happen not to have left a trace ... it is obvious that
geological history is circumscribed in no way by what we think we know
about it."

3 1955, Invertebrates and geologic time scale, pp. 547-574 in Arie Polder-

vaart, ed. Crust of the earth (a symposium),: Geol. Soc. Am., Spec.
Paper 62, viii and 762 pp. "The elementary but fundamental Law of
Superposition, on which every- geologist depends for indispensable evidence
of the relative age of sedimentary deposits laid down bed on bed, could be
formulated by reasoning, without studies made first in the field. On the
other hand, the nature of fossil assemblages in any given layer is not
subject to prediction, at least not without much accumulated experience."

MORRISON, R. B., GILLULY, JAMES, RICHMOND, G. M., and HUNT, C. B., 1957,
In behalf of the Recent: Am. J. Sci., vol. 255, pp. 385-393. ". . . geologic
. . . time is subdivided on a purely arbitrary, subjective basis, largely
for convenience in classification of formations and depending upon the
amount of detail recognized in the rock-stratigraphic record. Duration
has never been a determining factor in defining time units, just as thickness
has not been in defining formations or other rock-stratigraphic units."

MOULTON, FOREST RAY, 1929, Thomas Chrowder Chamberlin as a philosopher:
J. GeoL, vol. 37, pp. 368-379. "The methods of science, like styles in
dress, appear to come and go in cycles. At one period it is the fashion to
emphasize the accumulation of observational data and to disparage at-
tempts at their interpretation . . . The opposite extreme is the fashion
of hastily constructing theories, one after another, for explaining each new
phenomenon that is observed ... the fashion during periods of frenzied
research when priority is more highly prized than soundness and per-
manence."

MULLER, ARNO HERMANN, 1951, Grundlagen der Biostratonomie: Deut. Akad.
der Wiss., Berlin, Abh., Kl. Math. u. allgem. Naturwiss., Jg. 1950, Nr. 3,
147 pp. In seeking to reconstruct the sequences of events which have led
to the preservation of organic remains as fossils, biostratonomy employs
the principle of actualism.

MUNITZ, MILTON KARL, 1957, Space, time and creation, philosophical aspects
of scientific cosmology: Glencoe, 111., The Free Press and The Falcon's
Wing Press, x and 182 pp. "... the Perfect Cosmological Principle, like

BIBLIOGRAPHY 303

the Principle of the Uniformity of Nature, functions not as a factual state-
ment at all, capable of serving as a premise in an argument, but as a defi-
nition that functions as a criterion or rule of what in the language of science
is to be regarded as a law."

MURGHISON, RODERICK IMPEY, 1839, The Silurian System, pt. 1: London,
John Murray, 576 pp. "... while I rejoice in what I would call the
'Lyellian method' of testing geological phenomena by modern analogies,
I do not believe in the doctrine, that the dislocations of the present day are
produced by causes of the same degree of intensity as those of which geology
affords the proof. I must always be of opinion that, although they may
belong to the same class, the geological catastrophe (such as the overturning
of a mountain chain) and modern earthquake cannot be placed side by side
without our exclaiming 'sic parvis componere magna.'"

NAGEL, ERNEST, 1952, Some issues in the logic of historical analysis: Sci.
Monthly, vol. 74, pp. 162-169. "The distinction between history and
theoretical science is ... analogous to the difference between medical
diagnosis and physiology, or between geology and physics. A geologist
seeks to ascertain, for example, the sequential order of geologic formations,
and he is able to do so by applying various physical laws to the materials
he encounters; it is not the geologist's task, qua geologist, to establish the
laws of mechanics or of radioactive disintegration that he may employ."

NEWELL, NORMAN DENNIS, 1956, Catastrophism and the fossil record: Evolu-
tion, vol. 10, no. 1, pp. 97-101.

2 1959, The nature of the fossil record: Am. Phil. Soc., Pr., vol. 103, no. 2,

pp. 264-285. "Paleontological exploration of the past is a sampling pro-
cedure in which provisional estimates of the whole are made from small,
frequently biased, samples."

3 1959, Adequacy of the fossil record: J. Paleont., vol. 33, no. 3, pp. 488-499.

"... the limitations of the fossil record are not so much a matter of poor
preservation or insufficient quantity but rather insufficient collecting and
inefficient methods of preparing fossils."

4 1962, Paleontological gaps and geochronology: J. Paleont., vol. 36, no. 3,

pp. 592-610. ". . . the geologic systems, series and stages are all, in the last
analysis, world-wide units that are defined and identified by means of
their fossils. The time scale that is based on them is in no sense subjective
or arbitrary."

NICHOLSON, HENRY ALLEYNE, 1872, Contemporaneity of strata and the doc-
trine of geological continuity: Canadian J., n.s., vol. 13, pp. 269-281.
"... in so far as we can judge from the known facts of the present distribu-
tion of living beings, the recurrence of exactly the same fossils in beds far
removed from one another is prima facie evidence that the strata are not

304 CLAUDE C. ALBRITTON, JR.

exactly contemporaneous; but that they succeeded one another in point
of time, though by no long interval geologically speaking."

NIGGLI, PAUL, 1949, Probleme der Naturwissenschaften, erlautert am Begriff
der Mineralart: Basel, Birkhauser, \Vissensch. u. Kultur, vol. 5, xii and
240 pp. "Die Natur ist also nach gewissen Grundprinzipien gestaltet, und
es ist diese Gestaltung und Gliederung, die trotz der offensichtlichen 'Uner-
schopflichkeit' den Versuch der Naturerkenntnis zu keinem hofFnungslosen
Beginnen stempelt."

NOLKE, FRIEDRICH, 1937, Astronomic und Geologic: Deut. Geol. Ges., Zs.,
vol. 89, no. 3, pp. 167-175. Astronomical hypotheses, such as those related
to the secular decline in solar radiation and to the migration of the moon's
orbit away from the earth, have important implications for geologic his-
tory, e.g., warmer climates and stronger tides in the geological past.

NOLAN, THOMAS B., 1962, Role of the geologist in the national economy: Geol.
Soc. Am., B., vol. 73, no. 3, pp. 273-278. To problems involving the wel-
fare of nations, geologists might usefully apply the habits of thought in-
grained by the practice of their science. These include the tendency to
think in three dimensions, an almost instinctive use of the fourth dimen-
sion of time, an appreciation of the inevitability of change, and an accept-
ance of the variety of natural materials which lie below the surface of the
earth.

NORTH, FREDERICK JOHN, 1933, From Giraldus Cambrensis to the geologic
map, the evolution of a science: Cardiff Naturalists' Soc., Rept. and Tr.,
vol. 64, pp. 20-97. "The spirit of caution that was evident in Kidd's lec-
tures was also shared by those who founded the Geological Society of
London, for their chief aim was to observe and to record; they discouraged
further attempts at solving problems that could not be solved until more
information was available, and they would have nothing to do with hy-
potheses like those in which the previous century had been so prolific.
This attitude profoundly affected the course of geologic investigation,
which for many years consisted solely in the collection and classification
of specimens and facts."

2 1934, From the geological map to the Geological Survey, Glamorgan and

the pioneers of geology: Cardiff Naturalists' Soc., Rept. and Tr., vol. 65,
pp. 41-115. "Geology cannot be regarded as an exact science, for, from
the nature of things, the unknown must always be greater in extent than
the known. We can never hope to examine more than an exceedingly
small proportion of the rocks of the earth's crust, and all the fossils we can
ever hope to unearth will represent but an infinitesimally small part of
the whole pageant of life."

BIBLIOGRAPHY 305

OPIK, ERNST JULIUS, 1954, The time scale of our universe: Irish Astron. J.,
vol. 3, no. 4, pp. 89-108. "We will be guided by the principle of minimum
hypothesis, or economy of thought, which requires that new laws of nature
must not be used for the explanation of phenomena which can be ac-
counted for by known laws."

ORCEL, JEAN, 1954, Essai sur le concept d'espece et les classifications en min-
6ralogie et p6trographie: Soc. Franj. MineY. Crist., B., vol. 77, pp. 397-
432. The evolution of the concept of mineral species reflects an interaction
of discoveries and ideas developed in crystallography, chemistry, and min-
eralogy. According to the present concept, mineral species are differenti-
ated primarily on the basis of two criteria: crystalline structure, as deter-
mined by the refraction of x-rays, and chemical composition.

OSBORN, HENRY FAIRFIELD, 1904, The present problems of paleontology: Pop.
Sci. Monthly, vol. 66, pp. 226-242. "Just as the uniformitarian method of
Lyell transformed geology, so the uniformitarian method is penetrating
paleontology and making observations of animal and plant life as it is
today the basis of the understanding of animal and plant life as it was
from the beginning."

PAGE, DAVID, 1863, The philosophy of geology; a brief review of the aim, scope
and character of geological inquiry: Edinburgh, William Blackwood and
Sons, ix and 160 pp. "The philosophy of our science is thus to believe in
the fixity and uniformity of nature's operations, and under this belief to
regulate all our methods and arrange our results."

PARKS, WILLIAM ARTHUR, 1925, Cultural aspects in geology: Nature, vol. 116,
no. 2916, pp. 432-435. ". . . uniformitarianism is being questioned seri-
ously from both the inorganic and the organic points of view. We are
swinging back to a conception of a milder catastrophism variously ex-
pressed as rhythm, diastrophism, and so on."

2 1928, Some reflections on paleontology: Geol. Soc. Am., B., vol. 39, pp.

387-402. "... the modern tendency is to regard the time factor as the
motif of geology that is, to regard geology, in the first instance, as history.
We speak . . . more and more of the sequence of events in time."

PASSENDORFER, EDWARD, 1950, O zasadzie aktualizmu w geologii (na margi-
nesie ksiizki L. Cayeux," Causes ancienneset causes actuelles en g6ologie"):
Wiadom6sci Muz. Ziemi (Polish Geol. Mag.), vol. 5, no. 1, pp. 63-70.
"... les ph6nomfcnes gologiques anciens sont dus aux mmes causes,
agissant selon les mmes lois qu'aujourd'hui. Les mmes causes dans les
mfemes conditions donnent les mSmes effets."

PENCK, WALTHER, 1953, Morphological analysis of land forms (Translated
from the German by Hella Czech and Katherine Gumming Boswell):

306 CLAUDE C. ALBRITTON, JR.

London, Macmillan, xiv and 429 pp. "This state of affairs forms the sub-
stance of the fundamental law of morphology: the modelling of the earth's
surface is determined by the ratio of the intensity of the endogenetic to
that of the exogenetic displacement of material."

PIATNITZKY, P. P., 1939, Sur les dfauts de la terminologie du Pr6cambrien:
Int. Geol. Cong., 17th., U.S.S.R., Rept., vol. 2, pp. 15-24. Many of the
names given to geological systems below the base of the Cambrian are
ambiguous, illogical, subjective or otherwise unscientific.

PLAYFAIR, JOHN, 1802, Illustrations of the Huttonian theory of the earth: Edin-
burgh, Cadell and Davies, and William Creech (Facsimile reprint with an
introduction by George W. White, Urbana, Univ. of Illinois Press, 1956,
xix, xx and 528 pp.). "Every river appears to consist of a main trunk, fed
from a variety of branches, each running in a valley proportioned to its
size, and all of them together forming a system of vallies, communicating
with one another, and having such a nice adjustment of their declivities,
that none of them join the principal valley, either on too high or too low
a level; a circumstance which would be infinitely improbable, if each of
these vallies were not the work of the stream that flows in it. 55

POINCARE, HENRI, 1913, The foundations of science (Translated by George
Bruce Halsted) in Science and Education, a series of volumes for the pro-
motion of scientific research and educational progress, J. McKeen Cattell,
ed., v. 1: New York, Science Press, 553 pp. "It is often said: Who knows
whether the laws do not evolve and whether we shall not one day discover
that they were not at the Carboniferous epoch what they are to-day?
What are we to understand by that? What we think we know about the
past state of our globe, we deduce from its present state. And how is this
deduction made? It is by means of laws supposed known ... So that if
the laws of nature were not the same in the Carboniferous age as at the
present epoch, we shall never be able to know it, since we can know nothing
of this age, only what we deduce from the hypothesis of the permanence
of these laws."

POPOV, VLADIMIR IVANOVICH, 1940, Protiv morfologicheskikh ustanovok v geo-
logii (Against morphological tendencies in geology): Sovet. Geol., no. 5-6,
(May-June), pp. 167-172. The view that geology is basically a morpho-
logic science, devoted to a study of external structural features and their
relationships, not only inhibits the development of the science but is con-
trary to Engels' dialectic principle of 'struggle of the opposites. 9

POPPER, KARL RAIMUND, 1957, The poverty of historicism: London, Routledge
and Kegan Paul, xiv and 165 pp. ". . . while the theoretical sciences are
mainly interested in finding and testing universal laws, the historical

BIBLIOGRAPHY 307

sciences take all kinds of universal laws for granted and are mainly inter-
ested in finding and testing singular statements."

2 1959, The logic of scientific discovery: London, Hutchinson, 479 pp.

"Thus we are led back, by our concept of simplicity ... to that rule or
principle which restrains us from indulgence in ad hoc hypotheses and aux-
iliary hypotheses: to the principle of parsimony in the use of hypotheses."

POTONIE, ROBERT, 1957, Vom Wesen der Geschichte der Geologic: Germany,
Geol. Landesanst., Geol. Jb., vol. 74, pp. 17-30. "Wir wollen in den
Schriften der Geologen und Forscher blattern, welch sich in einem hoheren
Sinne um die Geschichte unserer Wissenschaft bemuht haben; nicht nur
registrierend, sondern nach allgemeineren Erkenntnissen strebend, zu
hoheren Schluszfolgerungen."

PRUVOST, PIERRE, 1951, Les refuges de Phypothfcse en geologic, in Sciences de
la terre, XXI Cong. Internal. Phil. Sci., Paris, 1949: Paris, Hermann,
pp. 3-39. Scientific hypotheses, though born to die, have ways of living
on, perpetuated in handbooks, embedded in popular opinion, upheld by
scientific authority, or concealed by jargon. Geologic hypotheses, in par-
ticular, may seek refuge at the bottom of the ocean, in the depths of the
earth, or in the abyss of time.

RAGUIN, EUGENE, 1951, M6thodes d'6tudes des formations gSologiques anci-
ennes, in Sciences de la terre, XXI Cong. Internat. Phil. Sci., Paris, 1949:
Paris, Hermann, pp. 17-28. The principle of actualism is insufficient for
the reconstruction of Precambrian history. For explication of the older
terranes, studies of the spatial distribution of metamorphic zones, structural
analysis of igneous rocks, comparative studies of orogenic belts, and geo-
chemical investigations are becoming increasingly useful.

RAMSAY, ANDREW CROMBIE, 1880, On the recurrence of certain phenomena
in geological time: British Assoc. Adv. Sci., 50th Meeting, Rept., pp. 1-22.
"... whatever may have been the state of the world long before geological
history began, as now written in the rocks, all known formations are com-
paratively so recent in geological time, that there is no reason to believe
that they were produced under physical circumstances differing either in
kind or degree from those with which we are now more or less familiar."

RASTALL, ROBERT HERON, 1943, Terminology in the geological sciences:
Nature, vol. 151, no. 3828, pp. 294-295. The decline of popular interest
in geology is partly due to "the volume and complication of the nomen-
clature now current ... By far the greater part of the trouble arises from
the activities of the palaeontologists ... It is commonly understood that
modern palaeontology is supposed to be founded on evolutionary lines,
but it really seems that the present tendency to hair-splitting distinctions

308 CLAUDE C. ALBRITTON, JR.

and infinite multiplication of new names is exactly contrary to the true
principle of evolution . . . these modern developments are mainly due to
museum specialists; people with the Civil Service mind, who spend all
their time indoors, surrounded by mountains of monographs instead of
mountains of rocks, and appear to delight in making everything as com-
plicated as possible/ 5

RAVIKOVICH, A. N., 1961, Uniformistskoye uchenie Layella i ego istoricheskie
korni (Uniformitarian teaching of Lyell and its historical roots): Akad
Nauk, SSSR, Inst. Geol., Ocherki po istorii geol. znanii, vol. 9, pp. 48-83.
In uniformitarian thinking, von Hoff and Hutton were the forerunners of
Lyell.

READ, HERBERT HAROLD, 1937, Certain aspects of metamorphic geology:
Liverpool Geol. Soc., Pr., vol. 17, pt. 2, pp. 103-114. "Time because it
means history, and in one aspect life, is the most fascinating aspect of
metamorphic studies. We shall some day read a long and complex history
at a glance in a single section of a metamorphic rock, and shall know even
the minor episodes of its dark and turbulent past."

2 1943, Geology in the war and in the peace: Junior Institution of Engineers,

London, J. and Record of Tr., vol. 53, pt. 7, pp. 179-187. "Any man
looking out of any window sees a geological laboratory in constant and
full-scale operation. 55

3 1957, The granite controversy: London, Thomas Murby, xix and 430 pp.

"The study of history of any kind depends upon documents and records.
For the history of the earth's crust, these documents are the rocks and
their reading and interpretation are often difficult operations. It is true
that for certain classes of rocks, made at the earth's surface, the uniformi-
tarian method is valid and sufficient . . . But many important rocks were
clearly made deep in the crust and no uniformitarian key can unlock their
secrets. 55

RENSCH, BERNHARD, 1960, The laws of evolution, pp. 95-116 in Sol Tax, ed.,
Evolution after Darwin, vol. 1 : Chicago, Univ. of Chicago Press. The
laws of evolution are of three kinds: laws of causality, laws of psychic
parallelism, and laws of logic.

RICKEY, JAMES ERNEST, 1952, Some aspects of geological research and their
practical application: Adv. Sci., vol. 9, no. 34, pp. 122-133. "A pattern
of evidence may include an anomaly at variance with apparently valid
conclusions reached on other grounds. Anomalies may not constitute vital
objections to a conclusion: on further investigation they may yield and
conform to the prior solution, or prove irrelevant. On the other hand the
recognition of anomalies has frequently been a most fruitful source of
discovery in geology as in other sciences. 55

BIBLIOGRAPHY 309

Rios, JOSE MARIA, 1962, Limitaci6nes y perfectibilidad permanente en la
cartografia geo!6gica. Problemas que plantean las tecnicas modernas:
Inst. Geol. y Minero de Espafia, Notas y Comuns., no. 65, pp. 127-138.
"For mi parte me alegro mucho de poder participar aun de la 6poca en
que la cartografia geo!6gica es un arte y no una fabricaci6n, en que el
g6ologo es una persona y no una maquina."

ROBERTSON, THOMAS, 1956, The presentation of geological information on
maps: Adv. Sci., vol. 13, no. 50, pp. 31-41. Geological maps are not real
portrayals of Nature. "At the very start of our field investigations we
abstract from geology as a whole the subject that we choose to study" and
the treatment of that subject will be "based upon a theory regarding the
succession and structure of the rocks." In the alternating flux of induction
and deduction attending the construction of a geological map, deductions
are sometimes built upon deductions "until an inverted pyramid of
formidable size is balanced upon a very small base."

RODGERS, JOHN, 1959, The meaning of correlation: Am. J. Sci., vol. 257, no. 10,
pp. 684-691. "... in stratigraphy the term correlation should and in fact
ordinarily does mean the attempt to determine time relations among
strata . . ."

ROEVER, W. P. DE, 1956, Some differences between post-Paleozoic and older
regional metamorphism (with discussion): Geol. en Mijnbouw, vol. 18,
no. 4, pp. 123-127. "There is apparently not only an evolution of life
during the history of the earth, but also some change in the character of
the metamorphic mineral assemblages produced during the main phases
of regional metamorphism of the various erogenic epochs."

ROGER, J., 1961, La documentation en g6ologie: B. Biblio. Fr., vol. 6, no. 1,
pp. 5-15. "Des progrfes consid6rables ont 6t6 r6alis& en geologic (au sens
large) depuis une trentaine d'annees dans les moyens et les techniques
d j observations. Par centre, dans la plupart des domaines, les gn6ralisa-
tions, les larges hypotheses, les theories n'ont pas sensiblement avanc6 et
souvent ne sont conserves qu'en raison de 1'absence de vues satisfaisantes.
Cette situation est due . . . pour une grande part & Pincapacit6 de r&mir,
dans un temps suffisamment court, une quantit considerable de donn6es
6parses dans une Iitt6rature 6norme ou issues d' observations nouvelles
qu'il fait souhaiter aussi abondantes et pr&ises que possible."

ROMER, ALFRED SHERWOOD, 1949, Time series and trends in animal evolution,
Chapter 7, pp. 103-120, in Glenn Lowell Jepsen, Ernst Meyr, and George
Gaylord Simpson, eds., Genetics, paleontology, and evolution: Princeton,
Princeton Univ. Press, xiv and 474 pp. "From time to time we are con-
fronted with ideological evolutionary theories, evolved in general in the

310 CLAUDE C. ALBRITTOX, JR.

philosopher's cabinet or theologian's study, which base evolutionary events
upon the 'design* of some external force a deity, or 'Nature' or upon
some mysterious 'inner urge 3 of the organism or its protoplasm ... In
general, the paleontologist can dispose of such theories by the application
of Occam's razor."

2 1959, Vertebrate paleontology, 1908-1958: J. Paleont., vol. 33, no. 5,

pp. 915-925. ". . . divisions between disciplines are tending to become
faint. Physics merges into chemistry; chemistry and physics merge into
biology and geology*; paleontologists, invertebrate and vertebrate alike,
bridge the gap between geology and biology."

ROSSITER, Mrs. P. M., 1937, Basic for geology, Psyche Minatures, gen. ser.
no. 90: London, Kegan Paul, Trench, Trubner, 164 pp. "There is no
danger now that the theories of geology will seem to be against true religion,
little that such forces as the vis plastica will be made the 'causes' of effects,
even less that the printed word will not get into the hands of workers in
far countries. The danger with which the present-day worker is faced is
that the printed word may not make sense when it gets there."

RUSSELL, ISRAEL COOK, 1907, Concentration as a geological principle: Geol.
Soc. Am., B., vol. 1 8 3 pp. 1-1 8. One of the important net results of physical
and chemical changes affecting the earth throughout geological history
has been the concentration of mineral substances which originally were
widely disseminated.

RUSSELL, RICHARD JOEL, 1958, Geological geomorphology: Geol. Soc. Am.,
B., vol. 69, no. 1, pp. 1-22. "Never before has the earth been so well
armored [by soils and vegetation] against processes of weathering and
erosion. Effects of this armoring must exist in sedimentary deposits to a
degree which makes it a bit hazardous to press too far any interpretation
of the remote past on the basis of the present."

RUTTEN, MARTIN GERARD, 1955, Mathematics in geology and the former ex-
tension of the Pre-Cambrian (with discussion) : Geol. en Mijnbouw, vol. 17,
pp. 192-193. "... every geological phenomenon is determined by an
almost immeasurable number of variables, horrifying in their complexity
and in the number of their interrelations. Every formula that uses a limited
number of variables is therefore but an extreme simplification."

2 1957, Origin of life on earth, its evolution and actualism: Evolution, vol.

11, pp. 56-59. "There was, of course, at one time a pre-actualistic period
in the earth's history . . . These early stages have often been called pre-
geologic. It seems better to speak of them as pre-actualistic, for a non-
actualistic early geologic history is well conceivable."

3 1962, The geological aspects of the origin of life on earth: Amsterdam, New

York, Elsevier, vii and 146 pp. "... the findings of geology are in complete

BIBLIOGRAPHY 311

agreement with the modern biological views on the origin of life through
natural causes."

SAINT-SEINE, PIERRE DE, 1951, Les fossiles au rendez-vous du calcul, in Sciences
de la terre, XXI Cong. Internat. Phil. Sci., Paris, 1949: Paris, Hermann,
pp. 78-84. Although generally regarded as a descriptive science, paleon-
tology is rapidly developing predictive capabilities.

SANDBERO, C. G. S., 1932, Der Grundsatz des Aktualismus und die Bestimmung
gewisser Ablagerungen als glaziogene: Deut. Geol. Ges., Zs., vol. 84, pp.
636-641.

SCHAFER, WILHELM, 1959, Elements of actuo-paleontology: Salt Marsh Conf.
Pr., 1958, Univ. Georgia, Marine Inst, pp. 122-125. Actuogeology and
actuopaleontology are based upon direct observation of events. "Because
the work is in the present there is the possibility of experiment. The result
is in all cases the knowledge of timeless laws. These laws in the hands of
the geologist and paleontologist cast light on the geological occurrences of
the earth's past, and they give an insight into the past life and death upon
the earth and into the environments of this life."

SCHAUFELBERGER, P., 1962, Geologic und Bodenlehre: Neues Jb. Geol. u.
Palaont., Mh., 1962, no. 4, pp. 179-203. Pedology draws its resources
from several of the physical and biological sciences, but geology by virtue
of its concern with problems of weathering in general is the discipline
which appropriately coordinates and unifies the various approaches to soil
science.

SCHEIDEGGER, ADRIAN EuGEN, 1960, Mathematical methods in geology: Am. J.
Sci., vol. 258, pp. 218-221. The mathematical method is indispensable in
testing the consistency of inferences drawn from different sets of geological
data.

SCHENGK, HUBERT GREGORY, 1961, Guiding principles in stratigraphy: Geol.
Soc. India, J., vol. 2, pp. 1-10. Compares original and modern usages
relating to the concepts of superposition, horizon tality, original continuity,
uniformitarianism, faunal succession, strata identified by fossils, rock-
stratigraphic unit, time-stratigraphic unit, and facies.

SCHINDEWOLF, OTTO EfeiNRiCH, 1944, Grundlagen und Methoden der palaon-
tologischen Chronologic: Berlin-Nikolassee, Gebriider Borntraeger, 152 pp.
Because fossils, taken in their natural order of succession, record an evolu-
tionary development which is both linear and irreversible, paleontology
provides the surest basis for geological chronology.

2 1944, Uber die Bedeutung der PalSontologie als geologische Grundwissen-

schaft: Reichsamt fur Bodenforschung, Jb., vol. 63 (1942), pp. 629-676.

312 CLAUDE C. ALBRITTON, JR.

3 1948, Wesen und Geschichte der Palaontologie; Probleme der Wissen-

schaft in Vergangenheit und Gegenwart No. 9, Gerhard Kropp, ed.:
Berlin, 108 pp. As a historical science, paleontology has provided an
empirical basis for evolutionary theory in modern biology. By the same
token, paleontology is basic to systematic stratigraphy and hence to geology
in general.

4 1957, Comments on some stratigraphic terms: Am. J. Sci., vol. 255, no. 6,

pp. 394-399. "The sedimentary rocks, by themselves ... do not yield any
specific time marks, setting aside the old law of superposition, which can
provide relative age indications only in a restricted manner and which is
unfit for age correlations."

SCHNEER, CECIL JACK, 1960, History in science, Chapter 10, pp. 159-185 in
The search for order; the development of the major ideas in the physical
sciences from the earliest times to the present: New York, Harper, xvii
and 398 pp. "The development of geological science, as a historical science,
within the larger framework of the growth of natural philosophy, has a
special importance that can be compared only with the development of
astronomy as a descriptive science."

SCHRODINGER, ERWIN, 1956, On the peculiarity of the scientific world view,
pp. 178-228 in What is life? and other scientific essays: Garden City, N. Y.,
Doubleday, viii and 263 pp. "This object, the vivid and colourful picture
of past events is one hundred percent purely ideal . . . Should one, there-
fore, omit it and study only the actual remains . . .?

The picture is not only a permissible tool, but also a goal ... it is hard
to understand why that which is self-evident in the historical sciences
ought to pass for heresy in the physical sciences, namely, to deal with
events and situations that are inaccessible to direct observation. The
historical sciences do that almost exclusively."

SCHUH, FRIEDRICH, 1937, Gedanken uber die bei der tierischen Entwicklung
hervortretenden Entwicklungsrichtungen: Palaont. Zs., vol. 19, no. 1-2,
pp. 116-126. From an evolutionary point of view, progress is a series of
psycho-organic victories over the inorganic world.

SCHWINNER, ROBERT, 1943, 1st die Geologic wirklich eine "historische" Wis-
senschaft?: Neues Jb. Min., GeoL, u. Palaont., 1943, Abt. B., no. 5, pp.
130-136. It is misleading to call geology an historical science. History
deals exclusively with man; geology is an explanatory natural science.

SCRIVEN, MICHAEL, 1959, Explanation and prediction in evolutionary theory;
Satisfactory explanation of the past is possible even when prediction of
the future is impossible: Science, vol. 130, no. 3374, pp. 477-482. "The
most important lesson to be learned from evolutionary theory today is a

BIBLIOGRAPHY 313

negative one: the theory shows us what scientific explanations need not do.
In particular it shows us that one cannot regard explanations as unsatis-
factory when they do not contain laws, or when they are not such as to
enable the event in question to have been predicted."

2 1961, The key property of physical laws inaccuracy, pp. 91-101 in

Herbert Feigl and Grover Maxwell, eds., Current issues in the philosophy
of science: New York; Holt, Rinehart and Winston, 484 pp. "The most
interesting fact about laws of nature is that they are virtually all known to
be in error. And the few exceptions . . . seem quite likely to become
casualties before long ... but they represent great truths so we forgive them
their errors."

SCROPE, GEORGE JULIUS BUNCOMBE POULETT, 1858, The geology and extinct
volcanos of central France, 2nd ed.: London, John Murray, xvii and 258
pp. "The periods which to our narrow apprehension . . . appear of in-
calculable duration, are in all probability but trifles in the calendar of
nature. It is Geology that, above all other sciences, makes us acquainted
with this important though humiliating fact."

SEARES, FREDERICK HANLEY, 1938, The concept of uniformity, growth and reac-
tions (Elihu Root lectures of Carnegie Institute of Washington on the
influence of science and research on current thought): Carnegie Inst.
Washington, Supp. Pub. 37, 50 pp. "Evolution . . . stands as the law of all
the laws of science and thus symbolizes two things of interest here: the
notion of pervasive recurrence and regularity in the world of phenomena
the uniformity of nature and the growth of the notion of uniformity itself;
in a word, the development of the idea that the phenomena of the physical world
can be described by simple laws"

SEDGWICK, ADAM, 1831, Address to the Geological Society, delivered on the
evening of the 18th of February, 1831, on retiring from the President's
Chair: Geol. Soc. London, Pr., 1826-1833, pp. 281-316. ". . . we all allow,
that the primary laws of nature are immutable . . . and that we can only
judge of effects which are past, by the effects we behold in progress . . .
But to assume that the secondary combinations arising out of the primary
laws of matter, have been the same in all periods of the earth, is ... an
unwarranted hypothesis with no a priori probability, and only to be main-
tained by an appeal to geological phaenomena."

SEEGER, RAYMOND JOHN, 1958, Scientist and theologian?: Washington Acad.
Sci. J., vol. 48, no. 5, pp. 145-152. ". . . all our views are based upon an
underlying assumption of the uniformity of nature, not only those views
of scientists but also those of theologians, for whom dependability is a
sine qua nan."

314 CLAUDE C. ALBRITTON, JR.

SEMPER, MAX, 1911, Bemerkungen iiber Geschichte der Geologic und daraus
resultierende Lehren: Geol. Rundschau, vol. 2, pp. 263-277. "Ausserdem
1st es in der Geologic gar nicht moglich, sich auf die Wiedergabe von Be-
obachtungen zu beschranken. Jedes Profil enthalt cine grosse Anzahl
von Angaben, die auf Schliissen, auf dem Zusammenwirken von Beob-
achtungen und theoretischen Elementen beruhen."

SEWARD, Sir ALBERT CHARLES, 1959, Plant life through the ages; a geological
and botanical retrospect: New York, Hafner, xxi and 603 pp. "It does
not in the least follow that because all living species of a genus are now
confined to areas with a certain range of temperature, therefore extinct
species, almost identical with the living species, were equally susceptible
to limiting factors."

SHERLOCK, ROBERT LIONEL, 1922, Man as a geological agent; an account of
his action on inanimate nature: London, H. F. and G. Witherby, 372 pp.
"Man's most permanent memorial is a rubbish heap, and even that is
doomed to be obliterated."

SHOTWELL, JAMES THOMSON, 1949, Time and historical perspective, in Time
and its mysteries, ser. 3, pp. 63-91: New York, New York Univ. Press
"... though we cover all time with numbers, we do so only in order to
find things in it, and to know where we are when we find them ... In
short we mark Time by events rather than events by Time."

SHROCK, ROBERT RAKES, 1948, Sequence in layered rocks; a study of features
and structures useful for determining top and bottom or order of succession
in bedded and tabular rock bodies: New York, McGraw-Hill, xiii and
507 pp.

SIGAL, J., 1961, Existe-t-il plusieurs stratigraphies?: France, Bur. Rech. g6ol.
et min., Serv. d'Inf. gol., B. (Chronique Mines d'Outre-mer, Suppl.
no. 297) an. 13, pp. 2-5. Distinguishes between faciostratigraphy, zoneo-
stratigraphy, and chronostratigraphy as aspects of stratigraphy whose
respective concerns are with the establishment of local stratigraphic
columns, the definition of units bounded by isochronous surfaces and the
understanding of earth history.

SIMPSON, GEORGE GAYLORD, 1941, The role of the individual in evolution:
Washington Acad. Sci., J., vol. 31, no. 1, pp. 1-20. "Whatever happens in
organic evolution, or indeed within the realm of the biological sciences,
happens to an individual."

2 1950, Evolutionary determinism and the fossil record: Sci. Monthly, vol.

71, pp. 262-267. "The fossil record is consistent with historical causation
that is in continuous flux, nonrepetitive, and therefore essentially non-
predictive."

BIBLIOGRAPHY 315

3 1950, The meaning of evolution, a study of the history of life and of its

significance for man: New Haven, Yale Univ. Press, xv and 364 pp. "It
used to be usual to claim that value judgments have no part in science,
but we are coming more and more to perceive how false this was. Science
is essentially interwoven with such judgments. The very existence of sci-
ence demands the value judgment and essential ethic that knowledge is
good."

A 1960, The world into which Darwin led us: Science, vol. 131, no. 3405,

pp. 966-974. Astronomy made the universe immense; physics and related
sciences made it lawful. "To all these discoveries and principles, which so
greatly modified concepts of the cosmos, geology added two more of funda-
mental, world-changing importance: vast extension of the universe in
time, and the idea of constantly lawful progression in time."

5 1960, The history of life, pp. 117-180 in Sol Tax, ed., Evolution after

Darwin, vol. 1: Chicago, University of Chicago Press. "All science is
philosophical, and the only philosophies capable of validation are those of
scientists ... A scientist cannot so much as make an observation without
reliance on a philosophical premise, such as the by no means self-evident
minimal premise that there really is something to observe."

6 1961, Principles of animal taxonomy: New York, Columbia University

Press, xii and 247 pp. "All theoretical science is ordering and if ...
systematics is equated with ordering, then systematics is synonomous with
theoretical science. Taxonomy, in any case, is a science that is most
explicitly and exclusively devoted to the ordering of complex data, and in
this respect it has a special, a particularly aesthetic . . . , and . . . almost a
superscientific place among the sciences."

SMALL, JAMES, 1952, The time scale of organic evolution: Irish Astronom. J.,
vol. 2, no. 1, pp. 21-26. ". . . we can recognize that there is, and has been,
as much law and order in organic evolution as there is and has been in
astronomical evolution and on a similar time scale of millions of years."

SMILEY, TERAH LEROY, ed. 9 1955, Geochronology: Univ. Arizona B., Physical
Sci. B., 2, 200 pp. Twelve essays on scientific methods which can be
applied to the dating of terrestrial events.

SMITH, JAMES PERRIN, 1900, Principles of paleontologic correlation: J. Geol.,
vol. 8, pp. 673-697. "The geological succession of faunas has some irregu-
larities and anomalies . . . but the displacements of the time scale are too
slight and the uniformity in various separated regions too great to lay
much stress on homotaxis as opposed to synchronism."

SOLLAS, W. J., 1900, Evolutional geology: Science, n.s., vol. 12, pp. 745-756,
787-796. With the growth of knowledge about the earth, catastrophic
geology was supplanted by uniformitarian geology, and this in turn by
evolutional geology.

316 CLAUDE C. ALBRITTON, JR.

BONDER, RICHARD AUGUST, 1956, Gedanken zur theoretischen Geotektonik,
pp. 381-395 272 Franz Lotze, ed., Geotektonisches Symposium zu Ehren
von Hans Stille: Stuttgart, Ferdinand Enke, xx and 483 pp. "Die in der
endogenen Geomechanik anzuwendende Methodik muss nach dem bekann-
ten Prinzip von c Versuch und Irrtum* vorgehen. Die verschiedenen Grund-
theorien sind fur den Theoretiker Probe-oder Testtheonen, die geotek-
tonischen Thesen sind die Testthesen**

SPENCER, HERBERT, 1884, The classification of the sciences, pp. 59-112 in
Recent discussions in science, philosophy and morals: New York, D. Apple-
ton. The sequence of concrete sciences astronomy, geology, biology,
psychology, and sociology forms a natural and intergradational series
which reflects the cosmic evolution of their subject matter.

SPIEKER, EDMUND MAUTE, 1956, Mountain-building chronology and nature of
the geologic time scale: Am. Assoc. Petroleum Geol., B., vol. 40, no. 8,
pp. 1769-1815. "The pattern of concepts, and the consequent way of
thinking that the geologist brings to his observations . . . inevitably will
control the data he obtains, either by selection or in outright discernment."

STENO, NICOLAUS (NIELS STENSEN), 1667, Canis carchariae dissectum caput
(The earliest geological treatise, translated from the Latin, with introduc-
tion and notes, by Axel Garboe): London, Macmillan; New York, St.
Martin's Press, 51 pp., 1958. "Since the ground from which the bodies
resembling parts of animals does not produce this sort of body in our time
(a); since it is likely that the soil in question has been soft in former times
(b), indeed, presumably has been mixed up with waters (c); why, then,
should we not be allowed to surmise that these bodies are remains of
animals that lived in these waters?"

STOREY, TARAS PHILIP, and PATTERSON, JOHN ROBERT, 1959, Stratigraphy
traditional and modern concepts: Am. J. Sci., vol. 257, no. 10, pp. 707-
721. "Whereas the traditional stratigrapher sets his sights high by regard-
ing attributes of time as the guiding factor in stratigraphy, modern workers
drift aimlessly about in their arbitrary philosophy that lithologic, biologic
and time criteria should be treated without simultaneously relating one
with the other."

STRAHLER, ARTHUR NEWELL, 1952, Dynamic basis of geomorphology: Geol.
Soc. Am., B., vol. 63, no. 9, pp. 923-938. "Two quite different view-
points are used in dynamic (analytical) geomorphology and in historical
(regional) geomorphology. The student of processes and forms per se is
continually asking, 'What happens? 9 ; the historical student keeps raising
the question, 'What happened?*"

2 1958, Dimensional analysis applied to fluvially eroded landforms: Geol.

Soc. Am., B., vol. 69, no. 3, pp. 279-300. "The rational phase of science,

BIBLIOGRAPHY 317

often associated with 'deductive science' in geology, consists of the formula-
tion of explanations and general laws through logical steps of reasoning
from a series of initial postulates which seem to be valid in the light of
prior observation and experience."

STUBBLEFIELD, CYRIL JAMES, 1954, The relationship of palaeontology' to stratig-
raphy: Adv. Sci., vol. 11, pp. 149-159. "Both palaeontology and stratig-
raphy are observational rather than experimental sciences, nevertheless
however accurately the observations are recorded, they are but retrospec-
tive in so far as they relate to natural processes."

SUZUKI, Kom and KITAZAKI, UMEKA, 1951, On the stratigraphical meaning of
time and space: Res. Inst. Nat. Res., Tokyo, Misc. Rept. no. 22, pp. 28-35.
The dualistic conception of stratigraphic classification, according to which
sequences of strata may be subdivided into lithological and chronological
categories which are independent of one another, reflects a misunder-
standing of what geologic time actually means. Since the chronicle of
earth history is based upon the arrangement of rock masses in space,
lithological and chronological units are conceptually inseparable.

TASCH, PAUL, 1954, Search for the germ of Wegener's concept of continental
drift: Osiris, vol. 11, pp. 157-167. Wegener's concept of continental drift
may have its roots in Plato's myth of the lost continent of Atlantis.

TEICHERT, CURT, 1958, Concepts of facies: Am. Assoc. Petroleum GeoL, B.,

vol. 42, no. 11, pp. 2718-2744.
2 1958, Some biostratigraphical concepts: Geol. Soc. Am., B., vol. 69, no. 1,

pp. 99-120. "Biochronology shows the order of succession in which events

occurred; a loosely fitting absolute scale comes from the radioactive

methods."

TEILHARD DE CHARDIN, PIERRE, 1951, La vision du pass6: ce qu'elle nous
apporte, et ce qu'elle nous enl&ve, in Sciences de la terre, XXI Cong.
Internat. Phil. Sci., Paris, 1949: Paris, Hermann, pp. 71-74. As our per-
ceptions of space are affected by the geometric laws of perspective, so our
perspectives of time are subject to certain natural constraints. In recon-
structions of the geologic past, the slow rhythms of the universe are re-
vealed, but the actual origins of entities which have evolved in time cannot
ordinarily be resolved.

THOMAS, GWYN, 1956, The species conflict abstractions and their applica-
bility: Systematics Assoc., London, Pub. no. 2, pp. 17-31. Discusses
differences among paleontologists, and between paleontologists and neon-
tologists, on the definition of species, and suggests that paleontological
taxonomy is, and will remain, as much an art as a science.

318 CLAUDE C. ALBRITTON, JR.

THORBURN, W. M., 1918, The myth of Occam's Razor: Mind, vol. 27, pp.
345-353. "The unfortunate carelessness of Tennemann and Hamilton . . .
has turned a sound rule of Methodology- into a Metaphysical dogma . . .
It is folly to complicate research by multiplying the objects of inquiry; but
we know too little of the ultimate constitution of the Universe, to assume
that it cannot be far more complex than it seems, or than we have any
actual reason to suppose/ 3

TIKHOMIROV, V. V., 1959, Aktualizm v trudakh russkikh geologov nachala
XIX veka (Actualism in the works of Russian geologists at the beginning
of the nineteenth century): Akad. Nauk. SSSR, Inst. Geol., Ocherki po
istorii geol. znanii, vol. 8, pp. 154-164. The uniformitarian, or formal,
approach to actualism leads to a dead end, because, according to this
approach, we cannot explain the origin of rocks whose analogues are not to
be found forming at the present time. On the other hand, the outright
rejection of actualism leads to agnosticism regarding the possibility of
deciphering the evolution of our planet.

TOMKEIEFF, SERGEI IVANOVICH, 1946, James Hutton's "Theory of the Earth"
(Hundred and fiftieth anniversary of the birth of modern geology) : Geolo-
gists' Assoc., London, Pr., vol. 57, pt. 4, pp. 322-328. "Towards the end
of the eighteenth century the heroic conception of human history was
being replaced by one of gradual development composed of the actions
and thoughts of a multitude of men and women. Condorcet was the expo-
nent of this idea and his book was published in 1794. To the heroic concep-
tion of human history corresponded the catastrophic conception of earth's
history. This Hutton replaced by what may be called gradualism."

2 1948, James Hutton and the philosophy of geology: Edinburgh Geol. Soc.,

Tr., vol. 14, pt. 2, pp. 253-276. Hutton's theory of the earth was based
upon the principle of the uniformity of Nature; the postulate of deter-
minism; the idea that the study of processes, rather than of matter, affords
the better approach to geological problems; and the principle of integra-
tion, which assumes that small events in their cumulative result lead to
great changes. Hutton's principal contribution to geology was his concept
of the geostrophic cycle, which he apparently borrowed from the life cycle
of organisms.

TOULMIN, GEORGE HOGG ART, 1783, The antiquity of the world, 2d ed.: London,
T. Cadell, xiii and 208 pp. "If something always has existed, or must have
been eternal, why not pay a deference to the magnificent and beautiful
objects of whose existence we are certain? Why not grant eternity to
nature? . . . Nature is invariably the same, her laws are eternal and im-
mutable . . ."

TOULMIN, STEPHEN EDELSTON, 1953, The philosophy of science; an introduction:
London, Hutchinson's University Library, 176 pp. "... in whatever

BIBLIOGRAPHY 319

sense we understand the Uniformity Principle, whether as assumption, as
discovery or as manifesto, it has one special weakness: that of irremediable
vagueness. A principle stated in such general terms can be of no practical
significance .... So it is not Nature that is Uniform, but scientific pro-
cedure; and it is uniform only in this, that it is methodical and self-cor-
recting."

2 1961, Foresight and understanding; an enquiry into the aims of science:

Indiana University Press, 116 pp. "Those who build up their sciences
around a principle of regularity or ideal of natural order come to accept
it as self-explanatory. Just because (on their view) it specifies the way in
which things behave of their own nature, if left to themselves, they cease
to ask further questions about it."

TRUEMAN, ARTHUR ELIJAH, 1948, Geology today and tomorrow: Adv. Sci.,
vol. 5, no. 19, pp. 184-193. Among the principal contributions of geology-
to general thought are the scale of geologic time, a world view of the
history of life against this time scale, and an appreciation of the insignifi-
cance of historical time as compared with the stages of human evolution
and of vertebrate evolution in general.

TSILIKIS, JOHN D., 1959, Simplicity and elegance in theoretical physics: Am.
Scientist, vol. 47, pp. 87-96. "Truth is usually simple, and the discovery
of simple equations for the description of natural phenomena gives some
assurance that simple results involve factual truths."

UMBOROVE, JOHANNES HERMAN FREDERIK, 1942, The pulse of the earth: The
Hague, Martinus Nijhoff, xvi and 179 pp. "The historical succession of
phenomena, their correlation and meaning, form the most attractive and
interesting feature of geology."

VOGELSANG, HERMANN PETER JOSEPH, 1867, Philosophic der Geologic und
mikroskopische Gesteinsstudien: Bonn, Max Cohen, 229 pp. "Wo die
Gegenwart aufhort, da fangt die Vergangenheit an; die eine kann nur
durch die andere erklart werden."

VYSOTSKII, B. P., 1961, Problema aktualizma i uniformisma i sistema metodov
v geologii (The problem of actualism and uniformitarianism and the system
of methods in geology): Akad. Nauk, SSSR, Inst. Filosofii, Voprosy
Filosofii, no. 3, pp. 134-145. Actualism and uniformitarianism are differ-
ent scientific concepts. Uniformitarianism is an hypothesis of a simple
cyclic development in Nature; actualism is a method of investigation, by
which the study of present phenomena and processes is used as a basis for
discovering the geologic past and predicting the future.

2 1961, Vozniknovenie uniformizma i sootnoshenie ego s aktualizmom (The

origin of uniformitarianism and its relation to actualism): Akad. Nauk,

320 CLAUDE C. ALBRITTON, JR.

SSSR, Inst. Geol., Ocherki po istorii geol. znanii, vol. 9, pp. 84-125.
"... as a rule, hypotheses and ideas never die; they are partly incorporated
into new theories . . . LyelPs periodicity does not contradict irreversible
evolution; it is only one of its aspects . . . The uniformitarian approach,
but not the uniformitarian philosophy, is valid on small scales, so that it
still remains significant in the same sense that the decimal scale remains
significant despite Einstein's theory of relativity."

WADELL, HAKON ADOLPH, 1938, Proper names, nomenclature and classifica-
tion: J. Geol., vol. 46, no. 3, pp. 546-568. "Geology is unfortunate in the
respect that the things pertinent to the subject are difficult to arrange into
a single well-constructed and complete classification."

WALTHER, JOHANNES, 1893-1894, Einleitung in die Geologic als historische
Wissenschaft: Jena, Gustav Fischer, vi and 1055 pp. "Wahrend die Geog-
nosie beschreibt, die Formationslehre ordnet, ist die Erdgeschichte eine
erklarende Disciplin, und daher bedarf die Geologie als historische Wissen-
schaft anderer Methoden und anderer Hilfswissenschaften, um ihr hohes
Ziel zu erreichen . . . Diese ... Art geologischer Erklarungs-versuche, die
mann gewohnlich als Aktualismus bezeichnet, wollen wir die ontologische
Methode nennen."

WATSON, DAVID MEREDITH SEARES, 1951, Paleontology and modern biology:
New Haven, Yale Univ. Press, xii and 216 pp. "Thus even the most
extended flights of morphological argument have led to predictions which
. . . have been verified by the description of actual fossil animals whose
structure conforms exactly to expectation."

WATTS, WILLIAM WHITEHEAD, 1911, Geology as geographical evolution: Geol.
Soc. London, Quart. J., vol. 67, pp. Ixii-xciii. "The History of the Earth,
so far as the Geologist is capable of following it through the geological
systems and formations, is a history of successive geographies, and of the
relations of these geographies to the living beings which successively
characterised them. In the reading and restoration of these geographies
there is but one unfailing guide, unceasing comparison, at every stage, of
the ascertainable geological phenomena of the past with the known geo-
graphical phenomena of the present."

WATZNAUER, ADOLF, 1956, Kritische Bemerkungen zur wissenschaftlichen
Begriflsbildung: Zs. angew. Geol., vol. 2, no. 2-3, pp. 64-65. "Das geolo-
gische Weltbild ist das Ziel der geologischen Wissenschaften."

WEBER, CHRISTIAN OLIVER, 1927, The reality of time and the autonomy of
history: Monist, vol. 37, pp. 521-540. "... suppose that we reconstruct
the world's past, by applying our present laws to the data of the world's

BIBLIOGRAPHY 321

present state. It is evident that we could never meet with a contradiction
in making this reconstruction, provided that no disharmony existed be-
tween our present data and laws. Then, suppose we find deep in the earth
a geological condition which shows a past different from the one we have
reconstructed? Will we conclude that the laws of mechanics have evolved,
and that they were different in the past? No, for the scientist can always
say that our present laws of mechanics are faulty, and must be modified
to cover the new facts. This amounts to saying that scientific laws do not
enable us to recover historical facts."

WEDEKIND, RUDOLF, 1916, t)ber die Grundlagen und Methoden der Bio-
stratigraphie: Berlin, Borntraeger, 60 pp. "Das grundlegende Prinzip
jeder Zeitmessung lautet: Die gleichen Ursachen bedurfen des gleichen %eitinter-
valls, um am gleichen Objekt die gleiche Veranderung hervorzurufen"

WEGMANN, EUGENE, 1950, Diskontinuitat und Kontinuitat in der Erdgeschichte;
ein Nachwort: Geol. Rundschau, vol. 38, no. 2, pp. 125-132. The tension
generated between catastrophists and uniformitarians continues to stimu-
late investigations and discoveries.

2 1951, L' analyse structural en geologic, in Sciences de la terre, XXI

Cong. Internat. Phil. Sci., Paris, 1949: Paris, Hermann, pp. 55-64. In
reconstructing a succession of shapes which have led to the development
of a geological structure in its present form, structural geology accomplishes
one of its main objectives without appeal to "causes" or "forces." This
manner of explanation is appropriate to geology, which is above all an
historical science.

3 1958, Das Erbe Werner's und Button's: Geologic (Berlin), vol. 7, no. 3-6,

pp. 531-559. "Die Herstellung der geologischen Geschichte ist eine Art
Zusammensetzspiel in Zeit und Raum. Viele Stucke dieses 'Puzzles'
fehlen und zwingen uns, um die Lucken herum zu bauen. Die Art des
Zusammenfugens der einzelnen Bausteine hangt weitgehend vom Leit-
bildeab."

WEIZSACKER, CARL FRIEDRICH VON, 1949, The history of nature: Chicago,
Univ. Chicago Press, 191 pp. "Since practically every event in nature
produces heat though often very small amounts every event is in the
strictest sense irreversible. Every pendulum comes to a stand-still. Even
the motion of the planets around the sun is constantly slowed down ever
so little by interstellar gas. Hence no event in nature is repeated exactly.
Nature is a unique course of events."

WESTGATE, LEWIS GARDNER, 1940, Errors in scientific method glacial geology:
Sci. Monthly, vol. 51, no. 4, pp. 299-309. "Geology differs from most
sciences in that its investigations are geographically determined ... It was
no accident that stratigraphic geology took its rise in central England . . .
and ... in the Paris Basin . . ."

322 CLAUDE C. ALBRITTON, JR.

WHEELER, HARRY EUGENE, 1947, Base of the Cambrian system: J. GeoL,
vol. 55, no. 3, pp. 153-159. ". . . how can we continue to indulge in a
concept which regards the base of the Cambrian system as a time horizon
but determines that base by physical criteria that are not amenable to
interregional correlation?"

2 (and Beesley, Edward Maurice), 1948, Critique of the time-stratigraphic

concept: Geol. Soc. Am., B., vol. 59, no. 1, pp. 75-86. "This variation in
age of lithogenetic units is ... a fundamental truth in stratigraphy, nearly
equal in significance to the laws of superposition and faunal succession,
and is ... designated therefore as the principle of temporal transgression''

3 1958, Time-stratigraphy: Am. Assoc. Petroleum GeoL, B., vol. 42, no. 5,

pp. 1047-1063. ". . . just as rock units of one kind or another are defined
to fill all space occupied by the determinable present geologic record,
time-stratigraphic units of one kind or another must be conceived or
designated to account for all interpretable space-time."

A 1959, Stratigraphic units in space and time: Am. J. Sci., vol. 257, no. 10,

pp. 692-706. "Time has no meaning in stratigraphy unless it is tied to the
spatial record and then substituted for its vertical dimension."

WHEWELL, WILLIAM, 1858, History of scientific ideas, being the first part of
the philosophy of the inductive sciences, 3rd. ed., vol. 2: London, John S.
Parker, xv and 324 pp. "I conceive that the assertion of an a priori claim
to probability and philosophical spirit in favour of the doctrine of uni-
formity, is quite untenable. We must learn from an examination of all
the facts, and not from any assumption of our own, whether the course of
nature be uniform."

2 1872, History of the inductive sciences from the earliest to the present

time, 3rd. ed., 2 vols.: New York, D. Appleton. "The effects must them-
selves teach us the nature and intensity of the causes which have operated;
and we are in danger of error, if we seek for slow and shun violent agencies
further than the facts naturally direct us, no less than if we were parsi-
monious of time and prodigal of violence."

WHITE, ANDREW DICKSON, 1930, A history of the warfare of science with theol-
ogy in Christendom: New York, D. Appleton, 2 vols. Chapter 5, "From
Genesis to Geology" sketches the contest between geological and theo-
logical views of nature to near the end of the nineteenth century.

WHITTARD, WALTER FREDERICK, 1953, The enigma of the earliest fossils:
British Naturalists' Soc., Pr., vol. 28, pt. 4, pp. 289-304.

WIENER, NORBERT, 1961, Newtonian and Bergsonian time. Chapter 1, pp.
30-44, in Cybernetics, or control and communication in the animal and
the machine, 2nd. ed.: New York, M. I. T. Press and John Wiley, 212 pp.

BIBLIOGRAPHY 323

"Bergson emphasized the difference between the reversible time of physics,
in which nothing new happens, and the irreversible time of evolution and
biology, in which there is always something new . . . The record of paleon-
tology indicates a definite longtime trend, interrupted and complicated
though it might be, from the simple to the complex."

WILLIAMS, HENRY SHALER, 1893, The making of the geological time-scale:
J. Geol, vol. 1, pp. 180-197. "Chronological time periods in geology are
not only recognized by means of the fossil remains preserved in the strata,
but it is to them chiefly that we must look for the determination and
classification on a time basis."

WILLIAMS, JAMES STEELE, 1954, Problems of boundaries between geologic
systems: Am. Assoc. Petroleum Geol., B., vol. 38, no. 7, pp. 1602-1605.
Discusses criteria that have been used to determine the boundary between
the Devonian and Carboniferous systems in Missouri, and concludes that
it is rarely possible to establish intercontinental correlation of thin units
near boundaries between systems.

WILLISTON, SAMUEL WENDELL, 1914, Water reptiles of the past and present:
Chicago, Univ. Chicago Press, vii and 251 pp. "... it is also a law in
evolution that the parts in an organism tend toward reduction in number,
with the fewer parts greatly specialized in function, just as the most perfect
human machine is that which has the fewest parts, and each part most
highly adapted to the special function it has to subserve."

WILMARTH, MARY GRACE, 1925, The geologic time classification of the United
States Geological Survey compared with other classifications, accompanied
by the original definitions of era, period and epoch terms: U. S. Geol.
Survey, B. 769, vi and 138 pp.

WILSON, JOHN ANDREW, 1959, Stratigraphic concepts in vertebrate paleontol-
ogy: Am. J. ScL, vol. 257, no. 10, pp. 770-778. "Type sections in strati-
graphic classification should have no more significance than name bearers.
They are one dimensional samples of three dimensional bodies of rock."

2 et a/., 1961, Geochronologic and chronostratigraphic units, Note 25 of the

American Commission on Stratigraphic Nomenclature: Am. Assoc. Petro-
leum Geol., B., vol. 45, no. 5, pp. 666-673. "... time can be logically
thought of, in geology, as being coincident with space."

WILSON, JOHN Tuzo, 1952, Orogenesis as the fundamental geological process:
Am. Geophys. Union, Tr., vol. 33, no. 3, pp. 444-449. "Now if, as is
believed to be the case, geologists accept the assumption that the general
laws of physics and chemistry apply to the earth and that no other ex-
clusively geological laws are necessary, then the fundamental processes

324 CLAUDE C. ALBRITTON, JR.

arrived at from physical reasoning must be the same as those arrived at
from geological reasoning."

2 1959, Geophysics and continental growth: Am. Scientist, vol. 47, no. 1,

pp. 1-24. "A belief that the earth does operate in a manner which can be
generalized in some simple way is justified because that is usual in nature."

WILSON, MORLEY EVANS, 1956, Early Precambrian rocks of the Temiskaming
region, Quebec and Ontario, Canada: Geol. Soc. Am., B., vol. 67, no. 10,
pp. 1397-1430. " c No vestige of a beginning 5 is as true of the earliest Pre-
cambrian rocks of the Canadian Shield as it seemed to Hutton of the
Highlands of Scotland in 1785."

WINTER, JOHN GARRETT, 1916, The prodromus of Nicolaus Steno's dissertation
concerning a solid body enclosed by process of nature within a solid: an
English version with an introduction and explanatory notes, with a fore-
word by William H. Hobbs: Univ. of Michigan Studies, Humanistic Ser.,
vol. XI, Contributions to the History of Science, Pt. II, pp. 165-283;
New York, Macmillan. "If a solid substance is in every way like another
solid substance, not only as regards the conditions of surface, but also as
regards the inner arrangement of parts and particles, it will also be like
it as regards the manner and place of production . . ."

WOLMAN, M. GORDON, and MILLER, JOHN P., 1960, Magnitude and frequency
of forces in geomorphic processes: J. Geol., vol. 68, no. 1, pp. 54-74.
"Analyses of the transport of sediment by various media indicate that a
large portion of the 'work' is performed by events of moderate magnitude
which recur relatively frequently rather than by rare events of unusual
magnitude."

WOODFORD, ALFRED OSWALD, 1935, Historical introduction to geology: Pan-
American Geol., vol. 64, no. 1, pp. 1-7.

2 1956, What is geologic truth?: J. Geol. Education, vol. 4, no. 1, pp. 5-8.

"... complicated problems can usually be broken down into simple parts.
Then each element can be attacked by one of several more or less rigorous
methods ... A solution, if found, may be (1) mathematically rigorous

(2) the sole surviving hypothesis, after exhausting other possibilities

(3) intuitive but fitting an extensive series of known facts, or (4) more or
less doubtful."

3 1960, Bedrock patterns and strike-slip faulting in southwestern California:

Am. J. Sci., vol. 258-A (Bradley Volume), pp. 400-417. When two slip-
solutions for faults are possible, "tentative choices may be made by use
of the rule Disjunctions minimae, disjunctiones optimae . . . This rule may be
considered a quantitatively parsimonious relative of Ockham's law . . ."

WOOLNOUGH, WALTER GEORGE, 1937, Fact and theory in geology, with special
reference to petroleum, salt and coal: Australian and New Zealand Assoc.

BIBLIOGRAPHY 325

Adv. Sci., Kept. 23rd Meeting, pp. 54-79. "I desire very diffidently to
suggest . . . that a set of physiographic conditions has existed from time
to time in past geological eras of which we have no true example at the
present time; and that variants of these conditions may serve to explain at
least some of the difficulties with which we are faced in an endeavour to
explain the origin of oil, salt, and coal."

WRIGHT, CHARLES WILL, 1958, Order and disorder in nature: Geologists'
Assoc., London, Pr., vol. 69, pp. 77-82. Geology comprises a group of
sciences unified chiefly in their concern with time and process. The
geologist should occasionally review his store of purely factual information,
for the order and regularity of nature are in the last analysis assumptions.

ZEUNER, FREDERICK EBERHARD, 1952, Dating the past; an introduction to
geochronology: London, Methuen, xx and 495 pp. Describes methods
and principal results of geochronology, "the science of dating in terms of
years those periods of the past to which the human historical calendar does
not apply."

Index To Bibliography

Abstraction

in scientific investigations: Niggli
Actualism (see also Uniformity, principle of)

Beurlen 1, 2; Bulow 2; Kumnierow; Rutten 2; Tikhomirov; Vysotskii 1
classical: Beringer, Hoff

compared \xith uniformitarianism: Hooykaas 1 ; Vysotskii 1
contemporary: Beringer

Engels' criticism of Lyellian uniformitarianism: Engels; Gordeev 2
essential to historical geology: Andree
Lomonosov's: Gordeev 2

principle: Backhand; Bubnoff 1; Cotta; Ebert; Hooykaas 2; Khain 1
relation to analogical reasoning: Hooykaas 1
temporal limitations: Bemmelen 1
working hypothesis: Bemmelen 1
Actuogeology

Schafer
Actuopaleontology

Schafer
Age determinations

iso topic: Kulp
Age of the Earth

geological evidence: Goodchild
Analogical reasoning
Gilbert 1, 2

basis of uniformitarianism: Hooykaas 1, 2
examples from historical geology: Watts
fallacy in organism-epiorganism metaphor: Simpson 1
in paleoecology: Cloud
in geomorphology: Baulig 3
limitations: Hooykaas 1
Applied geology {see Geology, applied)
Archeological stratigraphy

methods: Smiley
Astronomy

bearing on historical geology: Nolke
Atmosphere

preactualistic and actual istic: Rutten 3

326

INDEX TO BIBLIOGRAPHY 327

Biochemistry

comparative: Rutten 3
Biochronology

Schindewolf 1; Teichert 2
Bioecology

use of principle of uniformity: Clements
Biogenesis

multiple: Rutten 3
Biogeography

Mayr 1
Biostratigraphy

Moore 1

definition, contribution to geochronology: Teichert 2

nomenclature: Amer. Comm. on Strat. Nomenclature

principles and methods: Wedekind

synonym of stratigraphy: Schindewolf 4
Boundaries, chronostratigraphic

Bell
Boundaries, stratigraphic

difficulties of intercontinental correlation: Williams, J. S.

objective character of Mesozoic-Cenozoic boundary: Jeletzky 2

Cambrian

definition of base: Wheeler 1
Cartography, geologic (see Maps, geologic)
Case histories

Economic geology: Joralemon

Lake Bonneville Basin: Gilbert 1

value in scientific education: Gilbert 1
Catastrophes

climatic: King, C.

erosional: Wolman

extinction of animals, end of Mesozoic: Jeletzky 2

erogenic: King, C.

place in uniform nature: Dawson

volcanic: King, C.
Catastrophism

Cannon 2, 4; Eisley; Geikie 2; Gillispie; Greene; Le Conte; Wegmann 1; Wheweli 2

bibliography and quotations: Holder

catastrophist-uniformitarian debate, false issue: Herschel 2
heuristic values of: Wegmann 1

compared with uniformitarianism: Rutten 3

criticisms by German and French geologists: Bulow 2

Davis* tendency toward: Khain 1

defended: King, C.; Sedgwick

328 CLAUDE C. ALBRITTON, JR.

early Greek historiography: Collingwood 1

extinction of Mesozoic animals: Jeletzky 2

French: Billow 2

historically oriented theory: Bulow 2

illustrated by examples: Murchison

opposed: Ne\\ell 1

psychological roots: Berry 2

relationship to theism: Hooykaas 2
Categories

natural: Koch
Causality

Miller, H.

in geomorphology: Baulig 2
Causes

ancient vs. present in geology: Cayeux
Chronology

diastrophic: Schindewolf 1

geologic: Wilson, J. A. 2

ortho- and parachronology: Schindewolf 1

paleontologic: Schindewolf 1

Chronostratigraphic units (see Time-stratigraphic units, Time)
Chronostratigraphy

Sigal

boundaries: Bell
Civil service mind

workings in paleontology: Rastall
Classification

artificiality of geological schemes: Berry 2

biological: Simpson 6

concept of "type": Niggli

coordinate relational: Gilbert 1

crystals: Niggli

empirical- Gilbert 1

general: Koch

genetic classification in tectonics: Krynine

genetic, land forms: Davis 5

geological: Wadell

geotectonic elements: Krynine

land forms: Davis 5

linear relational: Gilbert 1

minerals: Niggli

natural classes: Gilbert 1

natural phenomena: Hedberg 1

paleontologic and neontologic: Hawkins 1

relational: Gilbert 1

stratigraphic: Interdept. Strat. Comm., USSR; Am. Comm. on Strat. Nomenclature

INDEX TO BIBLIOGRAPHY 329

Climatic fluctuations

basis of Pleistocene chronology: Brouwer
Columnar sections

asymmetry and transiiiveness: Griffiths
Comparative economic geology (see also Geology, applied)

case histories: Joralemon
Comprehensibility, hypothesis of

Schrodinger
Concentration of mineral substances

geologic principle: Russell
Continuity, principle of

Davis 2; Nicholson

use in petrology: Deer
Controversies

granite controversy: Read 3

uniformitarian-catastrophist debate: Cannon 1
Cope's law

Simpson 2
Correlation, stratigraphic

Berry 2

catastrophist influences: Berry 1

contemporaneity of strata: Huxley 1

criteria: Berry 1 ; Rodgers; Williams, J. S.

diastrophism an inadequate criterion: Spieker

difficulties of intercontinental correlation: Williams, J. S.

homotaxis vs. synchronism: Smith

meaning: Rodgers

paleontologic: Berry 1; Smith; Spieker

Precambrian: Haughton; James

principles: Berry 1

Wernerian influences: Williams, J. S.
Cosmology

cosmological uniformity: Bondi 2

functional and evolutionary aspects: Simpson 5

geological and biological implications of Milne's cosmology: Haldane 1

postulate of simplicity: Gold

problem of order: Simpson 5
Cosmolysis

principle of: Blanc
Covered-interval decision

Brown 2
Crystals (see also Mineralogy)

ideal and real: Niggli
Cuvier

contributions to systematic descriptive geology: Whewell 2
Cyclic phenomena

330 CLAUDE C. ALBRITTON, JR.

Gordeev 1

bearing on principle of uniformity: Barrell

earth history: Chamberlin 2

erosional: Bgthune 1, 2; Judson 2; King 1; Penck

geochemical: Earth

geologic: Umbgrove

Mutton's geostrophic cycle: Playfair; TomkeiefF 2

vs. linear processes: Schindewolf 1

Dacque's principle

Simpson 2
Darwin, Charles

debt to Lyell: Cannon 1
Darwinism

Beringer

implications regarding status of man: Simpson 4
Data, geological

categories: Krumbein 2
Davis, W. M.

contributions to philosophy of geology: Daly 2
Davis 5 cycle of erosion

Davis 5

critique: B6thune 2; Judson 2
Decision making

Brown 2
Deductive method

Popper 2

Davis* erosion cycle: Daly 2

explanation: Kitts

in geological mapping: Robertson

in geomorphology: Davis 5

weakness of Davisian geomorphology: B6thune 2
Deductivism

Popper 2
Definitions

Koch; Wadell
Demarcation

problem: Popper 2
Dendrochronology

methods: Smiley
Denudation

human vs. natural: Sherlock
Deperet's law

Simpson 2
Description

geologic: Koch

INDEX TO BIBLIOGRAPHY 331

Determinism

Chamberlin 3; Kober; Simpson 1

in evolution: Simpson 2

postulate: TomkeiefF 2
Developmental ism

Eisley
Dialectic materialism

Khainl

basis in history of nature: Engels

bearing on geology: Gordeev 1

laws: Engels

Lomonosov's: Aprodov

"struggle of the opposites": Popov
Diastems

BarreU
Diastrophism

aperiodic: Berry 2

geographically localized: Berry 2

laws: Bucher 3

not a reliable basis of interregional correlation: Wheeler 1

outgrowth of catastrophism: Wheeler 4

theory tested by paleontological evidence: Henbest
Dimensional analysis

applied to geomorphology: Strahler 2
Discovery, paleontologic

human factors: Henbest
Discovery, scientific

logic: Popper 2
Doctrine of irreversibility of evolution

Romer 1
Doctrine of permanency of ocean basins

Le Conte
Dollo's law

Simpson 2
Dynamics, geologic

scope: Whewell 2
Dynamism
Beringer

Economic geology (see Geology, applied)
Economy, law (see Simplicity, principle of)
Emergentism

Miller, H.
Empirical science

methodological unity: Hempel 1

qualitative and quantitative: Strahler 2

332 CLAUDE C. ALBRITTON, JR.

thought schemes: Koch
Empiricism

fundamental thesis: Popper 2
Endogenesis

Amstutz

geomorphologic: Penck
Engineering geology

scope: Galbraith
Entropy

landscape evolution: Leopold
Epigenesis

Amstutz
Epiorganisms

relation to totalitarianism: Simpson 1
Eras, geologic

origin: Stubblefield
Errors in scientific method

glacial geology: Westgate
Ethical systems

naturalistic: Simpson 3
Evidence

geologic: Richey; Vogelsang

paleontologic: Moore 1

paleoecologic: Cloud; Ladd

taxonomic: Simpson 6
Evolution, general

cosmic: Milne

nature: Kober

not a universal law of nature: Popper 1

organic and inorganic: Gotta

supreme law of science: Scares

universal phenomenon: Schuh
Evolution, inorganic

Goudge 2; Mason; Wilson, J. T. 2

geographical: Davis 5; Watts

geoxnorphic: Leopold

hillslopes: King 1

landscape: Crickmay; Judson 2

mineral assemblages: Roever

oceanic: Daly 1
Evolution, organic

Simpson 3

adaptation: Simpson 3

adaptative vs. progressive: Schuh

amoral: Simpson 4

archetypes: Simpson 1

INDEX TO BIBLIOGRAPHY 333

and catastrophism: King, C.

causality: Simpson 3

comparison of neo-Darwinian and orthogenetic theories: Grene

controlled: Haldane 2

Darwinian theories, reductivism in mechanistic aspects: Grene

doctrine of irreversibility: Romer 1

epistemological implications: Grene

evolutionary biology: Mayr 1

"explosive": Henbest

extinction: Simpson 3

fatalism: Simpson 1

finalist views: Simpson 3

historical process: Simpson 2

holistic views: Broom

human: Haldane 2

individualism: Clarke

laws: Hawkins 2; Rensch; Simpson 2, 3; Small

materialistic and mechanistic views: Simpson 3

metaphysical theories: Simpson 1

negative and positive eugenics: Haldane 2

neo-Darwinian dogma: Small

opportunism: Simpson 3

opposed to uniformitarianism: Gannon 1

organic and inorganic: Schuh

orthogenesis: Jepsen; Romer 1; Simpson 1, 3

paleontologic contributions: Simpson 4

plants: Seward

principles applied to civilizations: Hawkins

progress: Simpson 3

relation to Christian philosophy: Simpson 3

selection: Jepsen

slow development: Haldane 2

societal: Simpson 3

synthetic theory: Simpson 3

teleological theories: Romer 1

unknowable: Rutten 3

vitalist views: Simpson 3
Evolutionism

geoxnorphic: Khain 1
successor to uniformitarianism: Geikie 2
synthesis of uniformitarianism and catastrophism: Le Conte
Exogenesis

in geomorphology: Penck
relation to "higher superstition": Amstutz
Experimental geology
Bemmelen 1; Coombs; Demay; Leclercq

334 CLAUDE G. ALBRITTON, JR.

beginnings: Wegmann 3

geostatic and geodynamic aspects: Broggi

limitations: Bemmelen 2

misleading experiments: Kuenen

suitable problems: Kuenen

tectonics: Beurlen 2

theory of scale models: Hubbert 1

value: Kuenen

vs. natural science: Miller, H.
Explanation

causal: Gallic; Goudge 1; Popper 1

causal lines: Gallic

complex events: Kitts

deductive: Kitts

explanation sketches: Hempel 1

genetic: Beckner; Goudge 2

geological: Kitts; Vogelsang

historical: Beckner; Frankel; Gallic; Goudge 3; Hempel 1; Kitts

insufficiency of logical-empiricist views: Barker 1

logical character: Hempel 2

phylogenetic: Gallic

relation to description: Mises

retrodiction: Gallic; Gould

scientific: Hempel 1; Scriven 1

supernatural: Page

ideological: Beckner
Extinction

correlation with physical events: Camp

usual fate of species: Simpson 5

Facies, sedimentary

classification: Moore 2

concept: Gressly; Teichert 1

definition: Moore 2

faciostratigraphy: Sigal

nomenclature: Moore 2

variation in usage: Moore 2
Falsifiability

criterion for testing scientific theories: Popper 2
Fatalism

evolutionary: Simpson 1
Faunal sequence, principle of

Newell 4; Rutten

deduced from laws of superposition and organic evolution: McLaren

importance to concept of geologic time: Hawkins 1

importance to organic evolution: Hawkins 1

INDEX TO BIBLIOGRAPHY 335

law: Wheeler 2
Finalism, in evolution

Simpson 2
Formation, stratigraphic

ambiguities: Schindewolf 4

use of fossils in defining: McLaren
Fossil record

adequacy: Clarke

bias: Newell 2

causal gaps: Newell 2

fossil populations: Newell 2

imperfection: Hawkins 1; Lyell

nature: Newell 2

preservation: Newell 2

reliability: Newell 2
Fossils (see also Fossil record and Paleontology)

arguments for organic origin: Steno

evolution of ideas: Greene

history of interpretation: Haber

specimens, probability of discovery: Newell 3

use in defining formations: McLaren

Generalizations

distributive: Spieker

universal: Gould
Genetic sciences

Gallic

vs. nongenetic sciences: Miller, H.
Geochemical cycle

Earth
Geochronology

accuracy of radiocarbon dates: Hunt

astronomical aspects: Smiley

climatic fluctuations: Brouwer

definition: Jeletzky 1

importance of fossils: Schindewolf 4

methods: Smiley; Zeuner

principles: Brouwer

radiochemical dating: Smiley

Recent: Morrison

relative vs. absolute dating: Rutten 3

sedimentary vis-a-vis intrusive igneous rock: Rutten 3
Geochronometry

Kulp
Geocosmology

branch of historical geology: Broggi

336 CLAUDE C. ALBRITTON, JR.

Geodynamics

definition: Broggi
Geogeny

same as historical geology: Broggi
Geography (see Geomorphology)
Geological history

definition: Moore 2

record: Moore 2
Geologic climate units

definition: Amer. Comm. on Strat. Nomenclature
Geologic time scale (see Time, geologic)
Geologic-time units

definition, nomenclature, ranks: Amer. Comm. on Strat. Nomenclature
Geo-logismus

Kober
Geologists

classicists and romanticists: Bemmelen 1

contributions to society: Nolan

habits of thought: Nolan

increasing numbers: Nolan

intercommunication : Roger

motivation: Woodford 2

specialization: Nolan

Geology (see also Experimental geology,